Datasets:

64bits

/

lex_fridman_podcast_for_llm_vicuna

like 14

[ { "from": "human", "value": "But the point is they're all beautiful in their own little way and on all of them, especially if BN works. It's wonderful and it works fast. Simple thing, done right. Go to expressvpn.com slash lexpod for an extra three months free. This is the Lex Friedman podcast. Disappointed, please check out our sponsors in the description. And now, dear friends, here's Bert Kreischer. I read somewhere that you like Ernest Hemingway. Love him. There's a quote, there's nothing too writing. All you do is sit down at a typewriter and bleed. Do you agree with that?" }, { "from": "gpt", "value": "I agree. I agree. Well, I agree with that with journaling. I can't write stand up. I can't write a bit, but if I journal, I'll find something that I go just write. kind of writing where you're the pens moving faster than your brain and they're kind of like doing it together. The thing I liked about Ernest Hemingway, this is so stupid. I'm a little dyslexic and so I'm not a good reader. And so he wrote very simply and it wasn't until after I read a bunch of Ernest Hemingway, I was working in parts of Noble and this person said, don't you love his titles? And I was like, yeah, Sun Also Rises. And they're like, yeah, but And the sun also rises. And I was like, yeah, sun also rises. And they're like, no, no, the sun also rises. And I went, oh, oh, yeah, that is fucking badass. Yeah. And then I'm like always late to the party with anything. And then that night we were all doing coke. And it was like five in the morning and we ended up on the roof of my apartment in Greenwich Village. And it's Like I'm a sunset guy, I grew up in Florida, but sunrises in New York are electric blue, like electric blue. And I was like, the sun also rises." }, { "from": "human", "value": "Yeah, he was so good with just a handful of words, the simplicity, the choice of words. Like basically his process is carving down." }, { "from": "gpt", "value": "Right, drunk, edit sober." }, { "from": "human", "value": "Right, drunk, edit sober." }, { "from": "gpt", "value": "Yeah, but he was good at like, A small amount of words to get a point across. I'm the exact fucking opposite." }, { "from": "human", "value": "Well, he's also said the first draft of anything is always shit. Is that is that true for you?" }, { "from": "gpt", "value": "I think so. It's the best one I have." }, { "from": "human", "value": "Because Donald Hill from there. So for you, editing is destroying the original." }, { "from": "gpt", "value": "The first time I tell it, it's perfect. It's perfect it just it works and then i go great and then i take it and i try to fix it and make it better and jazzy i just joking razzle dazzle about my daughters and i drive and pass the deaf child sign. And i just told it on stage it just happened and louis was in the audience. He was, I love that joke. I said, really? He was, yeah, and I just told it and I went, done. Louis C. K. says he likes a joke. It's fucking, it's in the special. And I tried to fucking razzle, dazzle that joke into like a 15 minute bit. It got to the point where it was so, it was so bad that I told him, I did a podcast with him. I said, what did you like about that joke? Cause I can't get it to work anymore. Because it was just simple. It was simple that your daughter didn't know how they'd figure out where deaf kids lived. I went, that's it? He was like, yeah, I was like, nothing about gun control. And he was like, huh? I was like, oh, fuck, man. I really fucked that joke up. And then I had to, I had to go back to the beginning of it, but that first time that's when, that's when the joke comes out, purest for me. And then it's always chasing back to get that, get that first telling. Some jokes are different. Like the machine obviously was a much longer story and like, and like escape room for this one. The longer stories take a little work, but the, the simple ones, like, like the deaf child and pajamas and those first times are perfect." }, { "from": "human", "value": "What do you think is is it about the machine story that works that was so captivating for people?" }, { "from": "gpt", "value": "I think it's a I think I don't know. I really wish I don't know. I think it's a good story. It's a good story. I think everyone has something similar that happened in their life where they had a crazy night and that no one believed and they told their friends and their friend or they experienced that with a friend like that and And I think, I think that's gotta be it. I really don't know. I think there's a part of, I think there's a part of the, like our community of comedy fans that were that early Desquad that was fun to watch someone take something from the podcast and turn it into a bit on stage. I think there was a lot of that, but I think people identify with a wild, crazy night that got out of control that they've probably had too. I think I don't fucking know." }, { "from": "human", "value": "Do you think Putin knows who you are because of that story?" }, { "from": "gpt", "value": "I hope so. I fucking hope so. I hope so. But there's a part of me that you gotta understand is like, I'm not... I say stuff like that. I'm sure there's someone hears that and they go, who is this fucking narcissist? You're like, no, that's not it. I'm just telling you the truth. Like I understand what I'm supposed to say. Well, you know, I don't know. I don't know. That's hard to say, but I don't give a shit about that guy. I'm not that guy. I'm telling you everything that comes out of my mouth. The second it comes out, I'm going to be 100% honest. I don't know any other way to live. I kind of hope so. That would be fucking cool. And I look, I know he's a bad, I guess he's a bad dude. I don't follow politics done, but it'd be cool if like one day someone's like, uh, he's, he's on his computer and some guy goes, and he's like, and he's like, and he's like, and he shows it to him. And, and then he's like, I'm fucking this guy, huh?" } ]

[ { "from": "human", "value": "It's not out of the realm of possibility that Putin has watched it. And this podcast, this has kind of made me realize how small the world is." }, { "from": "gpt", "value": "I've gotten to meet some incredibly, like, interesting people. Like that's, I mean, that's the thing where I go. It's gotten really crazy. The internet's really insane and Netflix is really insane and the reach of like people that like this is really simple, but like I'm a big fan of the show Drink Champs. I didn't think in a million years Nori would ever know who I was. And then he offered me to come to his podcast and then when I got there, he told me he was a fan of mine and I went like, I remember moving to New York, listening to you and Capone like I'm sitting there going, wow. Again and then and the phone calls I've gotten since the movie came out and since razzle dazzle came out have been really insane when you realize. As much as I know about people and I like there's people I'm fans of that don't know who I am that like that are like, they're like some pitcher in fucking Florida that I follow the kids. I like I just fun to watch and pitch. There was a softball player in Tempe that she's do catching drills and I was really into softball with my daughter and we watch her catching girls. There's a girl deglazer who me and my daughters will watch her dance videos and then I hit her up. She's like, you know who I am. And you're like, oh, yeah, that's the way the world works." }, { "from": "human", "value": "What world leader would you drink with if you could see much in Putin? So Putin doesn't drink much. Yeah, Zelensky drinks more. Yeah, also not much." }, { "from": "gpt", "value": "Of course. He also would have been a good man. He would have been a fucking good one. I mean like I go Hitler was a pussy like he drank wine with it with a cut with a spoonful of sugar in it. You don't drink. What's Churchill? That's my hole and that's my guy. Yeah, you know, that's my guy, right? No, I don't know that's your guy." }, { "from": "human", "value": "So he, uh, he did champagne." }, { "from": "gpt", "value": "Well, you don't have to tell me a thing about Winston Churchill. He, I found out about his daily routine. Yeah. So his daily routine was a big breakfast, cigar, coffee, eggs over eggs, sunny side up, toast, bacon, sausage, tomato, beans, uh, fruit, orange juice, soft scotch, cigar in bed. And he'd sit there for three hours. He'd hold morning breakfast and that's they'd come in and he'd run through his day and then he'd get in the bathtub for fucking an hour and drink champagne And then he'd go to lunch and have some goose or whatever the fuck day back then with more champagne and then And then he'd take a nap, and like, I love that. So every year on the day of his death, and he's January 24th, I celebrate Winston Churchill's life by living the way he did. And so I have breakfast in bed with a soft scotch, and I just party all fucking day. That would be the guy to drink with. That would be the guy if you could pick anyone to drink with." }, { "from": "human", "value": "He drank every day throughout the day." }, { "from": "gpt", "value": "He lived like 94." }, { "from": "human", "value": "And was one of the most heroic figures in human history. Never, never surrender." }, { "from": "gpt", "value": "I mean, what a fucking badass thing. You know, he taught, he taught the queen. The queen was like, uh, like, uh, like a window licking idiot, right? Like she was, she had a third grade education. So like her dad dies and she's like 10 or 12 or whatever. The first thing they do is they pull her out of school. They're like, you've done learning. You're just going to be the queen. You gotta know, this is what you need to learn. Don't make, let people make eye contact with you. Put your hand out like this. Don't let people touch you. So she like, she had like a really, I saw the, by the way, I didn't do this research. I just watched the crown, okay? And Winston Churchill kind of taught her about world politics and was like, they like coming and she'd be like, Winston what did we do in Africa? He's like this is gonna be a tough one and so yeah, so but Winston Churchill's the badass But if I had a drink with someone living like drink there's like there's a too obvious one like obviously it's fucking Kim Jong-un right you kind of want to he drinks no he does You think my IVs are good in the morning, you know Kim Jong-un's IVs are next you're like I feel up and he's like it's infatamine He would be a badass to drink with." }, { "from": "human", "value": "I think he drinks hard liquor." }, { "from": "gpt", "value": "Uh, I think Scott Johnny Walker blue. That's what I've heard." }, { "from": "human", "value": "I don't know where you heard this, but I like this." }, { "from": "gpt", "value": "Okay. So I say things like that, but then I'm sure I, I, I, I don't know if you ever party with someone who just parties a little different than you. And then you notice like, like sometimes celebrity, like athletes and then like you're at a strip club and then they, they, they talk weird to a woman. You're like, Oh, that's not me. I'm so sorry. So that'd be the thing. The atrocities would be hard to get past with them." }, { "from": "human", "value": "Not just the atrocity, but the way of being. I agree with you. I actually don't like drinking with rich and powerful people often. There's something weird about them. Yeah. Not all of them." }, { "from": "gpt", "value": "And I don't like, I don't like a drinking meeting. Like when someone goes like, like people go like, Hey, uh, I'm taking you out tonight. We're drinking. Yeah. And then I go, Oh, that's not the way I like to do it. It gives me anxiety. Like, um, our lives are very different." } ]

[ { "from": "human", "value": "You and I?" }, { "from": "gpt", "value": "Yeah. Were you going to fraternity?" }, { "from": "human", "value": "No. We could start there yet. No, I was not. Okay. And the differences continue from there, I'm sure." }, { "from": "gpt", "value": "So when you get your pledge of pen, they take you out and their whole thing is you're going to, we're getting you fucked up. Yeah. That energy is not my energy. I don't like that. Like I just go, why don't we just get drunk and feel really good about ourselves? Like I never like took mushrooms to see Jesus. I just wanted to enjoy renex on a windshield. Like, so like that, that for me, uh, is like one of my things. So if Kim Jong-un flew me to North Korea to drink with him, I'd be like, I'd have to start really getting blackout drunk on the plane to enjoy drinking with him that day." }, { "from": "human", "value": "What's your favorite, like different occasions. What's your favorite drink?" }, { "from": "gpt", "value": "What's your favorite way of drinking? It's, it's, I've got as varies." }, { "from": "human", "value": "So what are the different contexts here? It's like one-on-one." }, { "from": "gpt", "value": "One-on-one?" }, { "from": "human", "value": "Like you sit down with, I don't know, Rogan invites you like it's a godfather to sit down and finally have a talk about something." }, { "from": "gpt", "value": "It's going to be whiskey. It's going to be like a nice whiskey. Neat. No, I like it on the rocks, not like lots of rocks. It won't be something expensive because Rogan is not, he's not a fancy boy, you know, like he's a real regular, he's a really regular dude. He's gotten less regular, the more millions of dollars he has. I think there's parts of him where you're like, like, where you're like, oh, wow, we can do this. Yeah. Um, but he's a rate, he'd be like Jack on the rocks or like, probably Buffalo Tracer, one of the, maybe a little whiskey he's buying. Rogan Beach, uh, whiskey rocks. And then like, that's a good, I haven't hung out with Joe by ourselves. I take that back. I was with him the other night, but like, it would be cool. My times with Joe's are always on podcasts. Yeah. And his I missed the times where it was we just have you and him at the store late night." }, { "from": "human", "value": "Yeah, I'm okay with podcast. I used to think this is like performative or something like this, but it's not really." }, { "from": "gpt", "value": "It's just a chance to really sit down. I enjoy it because of the people I could never sit down with. Like David Cross. I wouldn't never sit down with David Cross because we don't run in the same circles. But it's nice he has something to promote. And then he comes to my house and then we can sit down for an hour and talk. And then we're Kevin Smith. Like people I would never really hang out with. Tom. I enjoy being alone with Tom more. Off mic. Tom Segua. Yeah. We have a lot of history. There's a lot of things, secrets we know about each other, and there's a lot of secrets going on in each other's lives that I would never share on podcasts that I can talk to him about. And I really appreciate his, and I know this isn't on brand of us beating each other up, but I really appreciate his insight as a dad and as a dude and as a son. But he makes me giggle harder than anyone, so doing a podcast is fun." }, { "from": "human", "value": "So there's something on a podcast that's performative for you with him, and then there's a part." }, { "from": "gpt", "value": "With me and him, because we're trying to make each other laugh. But I'll tell you, if we're going back to drinks, I'll tell you, like a great afternoon drink is a Campari Spritz. It's one of my favorite drinks in the world. I don't even know what that is. Campari Spritz. Campari is an interesting alcohol Because I think it's a bitter or it's a liqueur and it's got a weird, rooty, earthy taste. It's red, so it looks like Kool-Aid. That's why I got into it. I thought Campania Nice would be oh, so nice. It's an old commercial, but man, it's fucking like sucking the green giant's dick. It is earthy. What's earthy? It's like, it tastes like roots. Oh, nice. It's not, it's not, it's, oh, there's a bitterness and like, it's just not great. And then someone introduced me to a Negroni which is Campari and gin and a little soda water and then Campari Spritz. Campari Spritz is Campari, Prosecco and soda water and like a wine glass and it is so enjoyable. It's got so many different profiles to it where you get the bitterness, but the sweetness of the prosecco and then it flattens it out with the soda water and it looks like goulade. So I love it. That's one of my favorite drinks to share in what context. Oh, you're in Italy. And you've been walking around Venice and everyone went their own ways and the whole family comes back together and you meet at a statue and your wife's sweating and she's uncomfortable and your daughters are 16 and 18 and they're like, what are we doing with the rest of the day? And you go, hey guys, it's Europe. Does anyone want to come party spritz? And everyone goes, I don't know what that is and you go, for compari spritz. It's minor in alcohol, so it's not gonna light them up, but it gives them a little bit of a buzz where they're like, dad, these are nice. Isla didn't drink hers, so I drank hers. Georgia had hers, Leanne had hers, and we ordered another round. I ordered, and they're fucking refreshing, and the buzz is perfect. It's not too much, because it's just a little bit of champagne and some compari. If the buzz is perfect, and then a couple of them, and then they kind of disappear, and then you just walk around again. That's a great fucking drink." } ]

[ { "from": "human", "value": "You have a drink alone? So there's two kinds of alone. One is alone, like at a bar. There's other people, but they're strangers, right? They're kind of like out there. Can't have done that. Keep going." }, { "from": "gpt", "value": "Keep going. Don't walk me through all the alone." }, { "from": "human", "value": "And then I'll tell you the rest. There's all kinds of... There's a alone on a train. Like transporting someone." }, { "from": "gpt", "value": "A train." }, { "from": "human", "value": "And then there's just a loan in an empty home." }, { "from": "gpt", "value": "Or a hotel room. Hotel. And it's Sunday night. Sunday night. And you're packing your bag and you got a flight at 6 a.m. Yeah. but road house is on and you're like, shit. So you get out of the front desk and go, is the bar still open? And they go, yeah, can I order some to take to my room? Sure. How about six Heineken's? Six Heineken's. Yeah, my wife wanted one too. She's not up there. You murdered her. And you just say to yourself, I'm just getting six so that if I have four, I just know I have two extra ones. And then you go through six and you're like, oh man, I'm not done yet. Dude, I've drank in every alone scenario you can possibly think of. I've drank, I've drank alone in a helicopter with a dude who didn't know I was drinking. He was like, I'm drinking here. And he's like, absolutely not. And then I poured it in a flask and opened it and drank it. And he's like, dude, we're in a fucking helicopter. You don't think I spell that? The best alone drinking you can ever do, in my opinion, now we're getting to the weeds on it, is alone behind someone's back. What do you mean? When they don't know you're drinking, but you're drinking. Like Christmas shopping. And your wife says, all right, let's all split up. And you go, cool. And it's like, it's like 11 o'clock on a Sunday. And they just opened that bar by the elevator in the Beverly center. And you just sneak over and you go, Hey man, can I get a double jack on the rocks, rocks, rocks? And they're like, sure. And you just have it. And you just go, let's just, and then you're off. And then you're like, yeah. And then you just little sneak one, sneak it, sneak it. And then you're buzzed and nobody knows. And then your wife's like, wait, have you been drinking? You're like, Yeah, I'm drinking by myself in a fucking mall on Christmas. How sad do you think I am? And then you go, yeah, I am." }, { "from": "human", "value": "What about mood? Do you ever drink in a dark place?" }, { "from": "gpt", "value": "So I broke, I had a chick break up or a chick cheat on me. When I was when I came back from Russia and she's something my best friend and I didn't know how to deal with the emotions I had and my buddy Mike Osborne said I can tell you this if you drink they go away and I went okay and that was the first and kind of last time I ever did that because I did it I've done it I've done it a couple times but it I try to avoid, I try to avoid if there's a motion going to alcohol, like meaning, not when it's anxiety, but like depression. I kind of sit in the depression because, or anxiety, I lean towards alcohol, like anxiety, like about flying or, or, or just like getting worked up on over something. But with depression, I try to avoid alcohol and just sit in it. Um, because I, and because I've gone there before, I've, but anxiety is a, I land, I land all the time." }, { "from": "human", "value": "But depression, heartbreak, all that, you just sit in the feeling." }, { "from": "gpt", "value": "Yeah. Heartbreak is a weird one for me. So like the last time I got my heart broken at Leanne dumped me, um, we were dating for like four months and she broke up with me. And my instinct was to have a drink. My friends went to this, uh, this, uh, Middle Eastern restaurant and I told him what was going on. And my buddy of course like, let's get a cocktail. And I was like, I don't want one. I got to figure this out. Cause I want this chick. And if I start drinking, I'll be like, ah, fuck it. So I got to figure it out. Like I wonder, I'm going to find out my dad dies one day or my mom, but my dad will be the rough one. My mom will be equally as rough. My dad and I were really close. And I wonder if I've thought about this a lot. If I'll have a drink dealing with that. Like I think about that a lot. I keep saying everything I get with my dad is extra. My dad, my dad had a 100% blockage in his widowmaker twice and both times they saved his life and he got a stint, I think 12 or 13 years ago and then he just got a stint this last year. But so everything I get to do, I feel like it's extra. And so, you know, he was just at the premiere of my movie and he got to see that and he had a very emotional response. And I got, I got that from him. And I was like, that's a little ice on the cake. I did, you know, the arena in Tampa was the first time he ever saw me do stand up. That was like a little extra. So all these things I'm getting with them are all extra. So I have a, you know, at least 74. So yeah, that'll be a tough one. At the premiere, he was, what, proud? Yeah, he came out, he was sobbing, crying. And he goes, I've underestimated you your whole life. And I was like, what? And he was shaking. He goes, I just underestimated you. I didn't know. I didn't know you were this, I didn't know whatever. It was really hard to understand him to be honest with you. I don't know totally what he said. He might have just not liked the movie." } ]

[ { "from": "human", "value": "I think you, uh, we're hearing what you wanted to hear. Uh, would you learn about life from your father?" }, { "from": "gpt", "value": "Uh, it's not going to make sense. I want to feel like sometimes I'm like a professional fucking wrestler because all I care about is you come and see me do stand up. So I, whatever I got to do to get you there, but the number one thing, my one takeaway from that man's humility and I applied, I've applied it in different ways throughout my career. On my 26th birthday, he gave me a really aggressive speech. He, I was hungover and I just thought he was going to say happy birthday. So I picked up the phone and I was like, Hey, and he was like, you are a tremendous piece of shit. And he broke me off. You have no humility. You have no this. You've never succeeded in anything. You're lying to yourself. You're lying to everyone around you. You're not doing what you're going after your dream. I broke you. I failed you as a father. This and that. I was like, I thought you're going to wish me happy birthday. And so in that speech, he told me if I wanted something, I had to go at it and I had to go after it as hard as I could. And I had to do whatever it could. I could do to achieve that. In this case, it was working, offering to do anything at the comedy club to get on stage. He told me, I told him, that's not how it works. And he said, no, that's how it works for some poor kid from Harlem. Because that kid needs it. He goes, you've grown up with privilege or life. You don't need it. You think you're going to be fine. And so I went and I ended up getting a job at the door, but I it was it was it was it's humbling. It's it's really like it makes me feel it made me feel uncomfortable. And so that humility is, it's the first time I really understood humility and I've applied that in many respects in like, in like getting to be a paid regular at the comedy store that happened late in life for me. And which I think most people would give up and be like, fuck that, I'm already a touring comic, I'm on TV, I don't need that. But it's something I wanted and I knew I wanted. So I just humbly went after it. And then I think I still apply it when it comes to selling tickets or selling a project to the fans. Or even to doing podcasts, I try to be as honest as I can about whatever I'm going through or whatever I'm dealing with. But when it comes to like selling things, I don't have a problem trying to make fun of myself or get in front of people so that they know that I have a show. Because the thing I'm good at, the one thing I know I'm good at, stand up. And I say that humbly, but like, I want them to come there. So like, I've heard people make fun of me like before they're like, yeah, I'm not some, I'm not some clown and a speedo who's one is going to, I'm a real comic. And I go, okay, but I want it. I want them to have my shows and you can be that guy and I'll be this guy." }, { "from": "human", "value": "Yeah. The clown and the speedo." }, { "from": "gpt", "value": "Yeah, it's like what are you gonna do? Funny people make fun of me for taking my shirt off like I guess behind my back. No one ever does it to me like no one would ever Comics aren't comics aren't comics aren't that manly Yeah, yeah, but I always go like I always go like who what do you care? What am I doing? What do you care?" }, { "from": "human", "value": "And you love taking shirt off more than anything You know, when something becomes meme and becomes super popular, it becomes easy to not love the thing anymore, to get tired of it. You still love it?" }, { "from": "gpt", "value": "More than anything. I like being shirtless more than having a shirt. I'm very uncomfortable in a shirt. Like right now, and I have tactile issues, like I have legit tactile issues. I'm most comfortable in stretchy jeans with a loose fitting belt. And then these shirts exactly, but I like more like a v-neck. I can feel this on my neck right now, and I can feel it on my arms. And so I'll sit a certain way, because it just rubs me wrong. But you have more comfortable shirtless. I can never wear what you're wearing." }, { "from": "human", "value": "Yeah, well, this is my shirtless. Each of us have to find our place in this world in terms of fashion." }, { "from": "gpt", "value": "I would love to be able to wear what you wear, because it looks good." }, { "from": "human", "value": "It feels like an escape from convention, but you or me, both of us. Yeah. Cause this, nobody wears this unless you work security for somebody." }, { "from": "gpt", "value": "Yeah, but it looks good on you. Does it? I don't know. You know, I think people think, I don't like, they would say like, what, what if you lose weight? You can't take your shirt off anymore. I go, no, I'm going to take my shirt off. You fucking out of your mind. I'm going to take my shirt off all the time." }, { "from": "human", "value": "You mentioned early on heartbreak after the Russia trip with a girl who slept with your friend. Let's go back to Hemingway quotes. The best way to find out if you can trust somebody is to trust them. So to go all in with the trust. I love Hemingway so much. Have you been betrayed in your life? Yeah. What do you think about trust?" }, { "from": "gpt", "value": "It's an interesting question. Um, I've been betrayed in my professional life a couple of times by really close friends. And, uh, I'm not someone who can like, I have a mantra that, I mean, I'm, I don't know if people have heard it or not. And I hate, I hate sharing it, but it's, but it's true. The mantra is, so you think I'm weak. Like, do you think it's like, that's my thing. Cause I feel like people take advantage of me and I go, so you think I'm weak. You forget everyone that like, and I'm named the, the people that have betrayed me, I'm a larger man than all of them. And, and, but I, and so that's what really fucks with me is I go, so you think I'm weak. Like where I grew up when you do stuff, there was physical consequences. I grew up in Florida. Like I've been beaten up a bunch. I'm not saying I can fight, but like I'm not scared of getting beaten up. And the, the times that I've been betrayed, I, and, and even, I know, even like by Ari, whichever, people see that drugging as a betrayal. And I did for a period, but it was, it was only because it was Ari and he was my, one of my best friends. And me and him had to work through that. Cause I, I, I couldn't not love him. Like I couldn't not have him in my life. He met, he means a lot to me. He really does. He's a great friend to have. But he's also, I don't know what was going on with him at that time, but he made a bad decision and I had to forgive him for that because I knew he knew my mantra. I think I'm harder to trust people now. I'm not as apt to trust people at all. But there was a moment, I don't know if I've ever shared this and I don't know how this will sound, but there was a moment that information got leaked about a thing and all my managers and agents, both, both, Tom and our managers and agents would like to either separate us or one of them get both of us. It's just more lucrative for them. So they have always kind of pitted us against each other a little bit and they do it subtly. They'll do it to me and my team and then his team will do it to me and then my team will do it to him and his team will do it to him. And there was a moment where information got leaked about this thing I was doing And my team was like, just so you know, your, your boy Tommy is, is the rat. And I had to, it was the one time when Leanne said, do you think Tom would do that? I said, uh, I don't know how faith works. Like I'm not a super religious person, but I will tell you that I believe in Tom and I don't think he did it. And I will stand by that ignorantly. And, and I will, and I will ride or die with him because I cannot live in a world where I don't trust him. Yeah. If I don't trust him, then I am so fucked. And, uh, and it wasn't him. It wasn't him. And it was, it's a cool feeling to know that you could like, that you could blindly trust someone, blindly have faith in someone, know that they're, they have your best interest at heart. But yeah, it took me a while to get there. Rogan helped me get there because I wasn't trusting him or Joey or Tommy or anyone. I wasn't trusting anyone. I'd gotten in a, Bad thing and I just wasn't trusting anybody. And Rogan was like, yo man, we're trying to be your fucking friend. And I was like, yeah, I don't, I don't need friends. I couldn't understand why Rogan was interested in me. And I said to someone, I was like, is he gay? Like, what the fuck? And then he was just a nice guy. He was just a nice guy." } ]

[ { "from": "human", "value": "That'd be hilarious if Joel was sexually attracted to you this whole time. This whole time is the reason he keeps inviting you back." }, { "from": "gpt", "value": "I would be more excited. I would be more excited if he tried to kiss me on the lips once. And just in a Scotty J way from Boogie Nights where he's like, what? Do you like my car?" }, { "from": "human", "value": "Yeah. You know, nobody would believe that story. No one. Yeah." }, { "from": "gpt", "value": "If I was Joe, I would kiss dudes all the time. And then when they didn't kiss back, you know, no one will ever believe you." }, { "from": "human", "value": "Yeah. When you trust people like that and they fuck you over, that can really, really, really hurt." }, { "from": "gpt", "value": "You end up on a treadmill in a hotel room fighting with them. You fight with them all the time. You fight with them all the time. You fight with them constantly. And I have this thing where I can ruminate on an idea, and I can't get it out of my head, and I hear the words they say." }, { "from": "human", "value": "So all in your head, on a treadmill, in your head. Just fighting." }, { "from": "gpt", "value": "Just fucking fighting, and then, yeah. And then one day you get past it. One day you just go, eh, fuck him. I'm not gonna leave them real estate and then you just forget about them. And then they reach out and they'll try to be a dick or a bully because they're bullies. They're bullies is what they are. Don't ever forget it. They're fucking bullies and they got bullied. It's not their fault. They got bullied. And then they try to bully you and you don't have any connection anymore. And you're like, oh, you can go fuck yourself. You can actually go fuck yourself. You're not my boss. You're not my wife. You don't fuck me. You don't pay me. Get the fuck out of my life." }, { "from": "human", "value": "But this kind of mantra of that you're not weak, You still want to be able to be fragile to the world and appear weak almost." }, { "from": "gpt", "value": "I am. I mean, that's the problem is I think I am pretty fucking weak. Like, I'm not the strongest dude out there. I have really thin skin. I get my feelings hurt all the time. Yeah. And so, but like, but it sucks when you watch like your friend see something in you that you go, or they go, fuck him. He's not paying attention. Fuck him. Let's just fuck him. Let's take his money. You know what, fuck him. He doesn't know what the fuck he's doing. Look, he's a drunk alcoholic and you go, oh no, bitch, I pay attention. Like I'm wide awake. I'm here. I'm working my ass off. You're not gonna fuck in." }, { "from": "human", "value": "Well, that's the tough thing about like you and Tom Segura. You're at the top of your game on top of the world, some of the most famous comics in the world. And you could see money start to creep in like business decisions. Yeah. I don't know. It's a tough, it's a tough place." }, { "from": "gpt", "value": "It's really tough. It's really tough for, you know, especially as we're growing. And, and, uh, I'll tell you what it, I don't know how much I'm allowed to say, but I can tell you this. We had a business deal where feelings were getting involved and money was involved in that. And the money was causing the feelings. And one of us said, to the other person, our feelings are more important than the money. So let's just split the money. And it was really like a solid moment where both of us were like, that is how we should run this." }, { "from": "human", "value": "That's the interesting things with bands. You have popular bands that split even. Even though it's clear that the contribution is not even. It's like there's a front man and all that kind of stuff, but it's ultimately the right thing to do to say, fuck you to the money for the most part and prioritize the feelings." }, { "from": "gpt", "value": "Yeah, especially because I couldn't do it without him. He couldn't do it without me. We do it separately. We'll have guest bears on. Some will blow up. Some won't. You can never figure it out, but when we're together, it does well. And so we need each other for it to work. And we've, you know, like I said, we're on a handshake deal with two bears and no one likes that. But we're like, Hey, I mean, we, I don't like, I, to the point where I like, he'll call me and go, yo, we have a deal to do this for X amount. Are you in? And I go, yeah. I don't check the books. I don't need to. I don't, I don't, I know him. I just know him." }, { "from": "human", "value": "Yeah. What do you love most about Tom Segura?" }, { "from": "gpt", "value": "We saw a gay couple one time. This is a good start. And there was one guy was in khakis and one guy was in jean shorts with a python around his neck. And Tom goes, if we were gay, which one would you be? I said, I'm fucking python. And he was like, yeah, I'm khaki. And I go, really? And he goes, yeah, that's why we work. And he goes, all relationships are for pythons and khakis. And I went, really? And he goes, look at you and Leanne. Who do you think you are? I go, I'm Python. And he goes, she's khaki. And he goes, me and push. I go, oh, she's Python. He goes, and I'm khaki. He goes, I'll always be khaki. And sometimes I'll have a relationship around the Python, but it's, I'm usually a khaki. And I was like, yeah, I am a little bit of a lunatic. I'll tell you the one moment that explains Me and him, the world will never see. It was just a great moment. We got stem cells with Rogan one weekend and one day. And I'd had surgery in my arm and I didn't want to do it. I was terrified. And Tom just said, just come out and meet up with us. Tom knew he was going to make me get them. And so I got them. And he, you know, Joe's there. And so we're just all talking. Yeah, yeah, yeah, yeah. And Tom can see that I'm spinning out a little bit. And I get in the car and Tom's got a Porsche. So we're like right next to each other. And I just want to feel if it's swollen and it is a little swollen because I just got an injection. And then in Tom's driving and he's got his hand on the thing. He just puts his hand on my arm and he goes, you're okay. He just, he was watching. He knows me so well that he knows that I'm spinning out of control. And he's like, you're okay. And then I was like, and I felt okay. I was like, you're right. I'm okay. What the fuck am I doing? My brain's that brain." } ]

[ { "from": "human", "value": "Man, that guy got so much love in him." }, { "from": "gpt", "value": "Dude, he does. He has, you know, he had a fantastic relationship with his dad. And I think whatever that dude did to create that dude, he hit it out of the fucking park because he has so much love that he doesn't need reciprocated like at all. Like he can just love And then he feels good for him to give and to like take care of someone. And like, you know, he started the whole birthday gift thing. Cause, and he didn't need to. He just was like, Hey man, I had this bike. I thought you'd like it. And then of course I've turned it into a fucking bit and now we're upside down. I'm 120 upside down on a fucking race car that we can't make our money back on." }, { "from": "human", "value": "I love it. Uh, what do you love most about Joe Rogan?" }, { "from": "gpt", "value": "Oh, that dude. I could go on and on about Joe Rogan. Joe Rogan is the most selfless individual I've ever met. He doesn't care if he has you on the podcast. He wants you to be the funniest person in the room. He wants you to be the smartest person in the room. Let me tell you something. Joe Rogan is 10 times smarter and funnier than me. He wants you to shine He wants to put you over to his fans. That shit did not exist before he started doing his podcast. People didn't share you with their fans. No one was like, hey man, come on on the road with me. They brought you on the road because they wanted you to suck and they put you in bad scenarios. Joe Rogan wanted you to shine. He gave so many people Beautiful career gave them careers gave them Joe could have looked at his podcast as his thing. This is my thing I bring on these guys. I'm gonna be the funniest kind of fucking room. I know some guys that have done that It might be me. It might be time I talk over people and not listen really well. Joe did the opposite. Now look, I've tried to learn from Joe on my podcast and do the same. That guy, when you think about it, you name 10 of his friends, I bet eight of them are doing arenas right now. They're all doing theaters. All his friends do theaters. All his friends do theaters. He is, he is just, you know, I, he, and he will accept this compliment. I tried paying it to him, uh, in, in Austin when I was in Austin last night. Well, he's not gonna accept any compliments. I gave him a gift too and I said, hey, I need to show you. I can't just tell you this because it's gonna fall in deaf ears. I need to give you a minute so you have something so you know that I think about you and I'm grateful for having you in my life a lot for many reasons. I have more friends in my life because of you. He taught me how to trust people again. I have a career because of you. I have a major motion picture because of you. I do arenas. Because of you without you in my life a lot of these things maybe never would have happened I never would have told the machine on stage I wouldn't have gotten I wouldn't never started a podcast I wouldn't have three podcasts like I mean all these things are things He not just he didn't just take me to the water one time and give me a drink He he gave me a machete and he said this is how I get to the water You got to put your own path, but just give me your heads up. You can also use my path whenever you want I mean, he, he is the most selfless individual. He will go down in history as one of the 10 most important people in stand-up comedy without a doubt. He changed the game. It didn't exist like this before him. And, uh, and he won't accept that compliment. What was the gift you gave him? I can't tell you. Is it a dildo? He was a dildo, double-sided dildo. I go, Hey, when you get free time, can we scissor fuck each other?" }, { "from": "human", "value": "Yeah. Uh, you're happily married, uh, for 19 years. So I think you're a good man to ask, how do you find love? How do you find, how do you date in order to find a love of your life, of a lifetime?" }, { "from": "gpt", "value": "This is going to sound. Off. But don't. Don't tie yourself down, meaning like, don't spend time. fucking with those little mini shrimp and peeling them every night when you know you're looking for a lobster. And so I think what happens with a lot of guys, they're so busy and just getting pussy and going from here to here to here. I was never that guy. I always wanted to find someone who understood me. I always wanted that. And so, granted, I haven't had sex with a lot of chicks, but that's And it's not by choice. I'm saying like maybe there was times I would have done it, but I know that I was always looking for someone to figure out to get me. And then I will say this, when you know, you know. When you know, you know. When you think there's, I can't imagine a life without this person. That that happened to me and I can only say my experiences I had Leanne I loved her and I and I didn't even realize it and the second she dumped me I went oh no no no no no oh I I don't get to live a great life without her she's the thing that's gonna give me a great life I knew it Undisputed I went for a run and she dumped me. I went to my shower. I cried. I went to Barnes and Noble. I bought men are from Venus women are from Mars I've watched Fight Club in my closet. Yeah, I didn't drink I drank a lot of Nyquil to sleep and then I went for a hike in Runyon Canyon I was running back down outpost by Ben Affleck's house and and the Sun was setting and I just said a God if you're listening Just give me this chick back. I'll never fuck it up. This is the one. I won't fuck it up. I won't cheat. I'm not going to be a bad dude. I want to have kids with her. I'll be there. I'll be a great fucking dude. I'll take her out on dates. Like, I'll love this chick. And I got home and she was in my room in a sundress. And I was like, oh, fuck you. I didn't know he was real. I was like, shit, man, give me like one more night. Let me get some strange tale real quick before I commit forever. But I think when you know, you know, this is so silly, but I follow that in business. Like when you know, you know, and like my assistants with us and someone goes, how did you, how do you find a good assistant? I go, when you know, you know, like people will show you who they are. Uh, I can tell you when I fell in love with my assistant, we were doing a TV show and, and, and we worked past dinner and all the dinner places were closed and he just said, Hey, I got like four meals. I didn't know which one you'd want. So I got four different meals for you. I ordered them. They're up in your room and just let me know whichever you want. I'll take care of the rest of them. Yeah. And I was like, oh, that dude thinks about other people before himself. He didn't even get himself dinner. He just got me four dinners. And I was like, fuck. I did the same thing with my social media manager. The way she was operating was I was like, I feel like she's thinking about me. And we connected on so many levels and so many levels, I could talk to her about things and ideas I had. And then I was like, okay, I want her. I want him. I want, I think you, I think it's gotta be that thing when you know, you know, and it can be so simple as like, like a personal trainer or like a jiu-jitsu class or like just where you go, oh, I'm connecting right now." } ]

[ { "from": "human", "value": "But you have to allow yourself to notice that person." }, { "from": "gpt", "value": "I think you do. I mean, I think you have to put yourself out there and you have to be available for it. You know, that's the hardest part is just being available for it. Cause so many people just want to be busy just dating. I just want to have someone in my life. I don't want to be bored. I don't want to, what if I die alone? I don't want to, I like, just be available for it." }, { "from": "human", "value": "Uh, what is it? You said peeling the baby shrimp. Yeah. When you could be having a lobster. Lobster is still a lot of work, I have to say." }, { "from": "gpt", "value": "Yeah, but, well, yeah, but it's lobster. This is a fucking baby shrimp. Or like, you ever get, so you ever get the ones in Brazil where you just go fucking, I'll eat the shell. I don't care anymore. When you're so fucking hungry, you just handful of lobster like it's popcorn, or maybe shrimp like it's popcorn. I've done that in Vietnam. I've done that where I just go, it's too much work." }, { "from": "human", "value": "So when you're in it, what's the secret to successful marriage? What do women want, Burt Cresher?" }, { "from": "gpt", "value": "Oh, fuck. I'm the wrong guy to answer that." }, { "from": "human", "value": "I think- How to make love to a woman by Burt Cresher." }, { "from": "gpt", "value": "How do you make love to a woman? I'll tell you how I do it. I go down on her first. Make sure she has an orgasm. Yeah. And then I get my turn and that's over pretty quick. It's like a, it's like at your ears pierced. It doesn't feel a little prick and it's over. I, um, I don't know if I've done a great job at keeping her happy. I think I keep her interested. Like, I think I, I, I. I keep her occupied like I'm like a little bit of a rodeo clown like I don't know. I mean like I know we're happy right now. I know we're really happy But I don't know if I ever did As good a job as she did like she's always been Like, she's always been a gangster. Like, did everything. Just does everything. She does it all. She does everything. Like, fucking everything. And she loves it. She just does it. And then she shows love for you by taking care of you. And so, like, I have a lot of time just been this, like, almost, like, step-and-repeat husband where it's, like, I come in to take a picture, like, hey, Monday, Tuesday. I'm on the road Wednesday. Take care, guys. Or, like, travel channel. I'd live for two weeks at a time. I left for the movie for three months. So, like, I don't know that I've ever done a bang up job. I buy it, like I go, I remember one time I was just like, I don't cheat on her. I don't hit her. I don't yell at her. I'm a pretty good fighter. I'm a really good fighter. Like, like I'm good at like going like, Hey, we're on the same team." }, { "from": "human", "value": "Oh, you mean de-escalation." }, { "from": "gpt", "value": "I'm really good at like, I'm very self, I'm very accountable. I'm very self-correcting. Like if we're in a fight and she points something out and I go, oh, you're right. I go, fuck, you're right about that. You're right. You're totally right. Oh, in the moment. In the moment, I'm really good at that. And my wife's not. She's not a good apologizer. Like she needs to sleep on an apology. And then she can do it. I can apologize in the moment if I realize I'm wrong. And I'm really quick to find my fault. I look for my fault because I go, tell me what I did. Adversely as a parent, I have pulled my daughters aside and been like, yo, that wasn't you. That was mom. I'm giving you a heads up. Cause I don't want, cause I got really confused as a kid. Cause my dad always made sure I knew I was wrong. And sometimes I wasn't. And then as a kid, you're kind of fucked up. And you're like, well, how do I know if I'm wrong? And I think that raises, so I was really good with my girls going like, go, and then, and then parenting Georgia taught me a lot about self-correcting." }, { "from": "human", "value": "Self-correcting." }, { "from": "gpt", "value": "What's that? My oldest. I'm only parented like my dad parented, like just like raise your voice, get up to here. That's it. No, that doesn't fuck this. How come the dogs don't have goddamn leashes on the colors on their necks? And so with Georgia, I had to learn through therapy that if the, if the, if the consequence is here and the highest it goes is here, then why the fuck wouldn't they lie and do everything? Cause they go, well, I know what it's like to hear, but if you can vary your consequence with a child, and find some conversation in it, then all of a sudden they're like, well, fuck. Well, I don't wanna lie, this does suck. I like this, I like where we talk. And he says, write a thousand words on why driving around your kids in a car, kids in a car is bad." }, { "from": "human", "value": "So threaten extreme violence every once in a while, just to... Yeah, let him know the dog can fucking bite." }, { "from": "gpt", "value": "The dog can bark the dog can bite, but if you pet him on his belly, it'll wag its tail. Yeah." }, { "from": "human", "value": "Uh, do you worry about being away too much? I mean, you're one of the most creative content pushers and creators out there. You're just constantly innovating, constantly putting stuff out there. But do you worry that pulls you away from like, uh, the mundane experience of life that can bring contentment and happiness?" }, { "from": "gpt", "value": "Yeah, I feel like sometimes it cannibalizes your real life, where you start going, you start thinking in promo videos, or you start looking at a vacation as a bit. We should go zip lining. Why don't you do, what shoes are you wearing? It can definitely cannibalize your life. My wife's really good at going like, no, like no phones. The girls are really good. The girls, when I started Instagramming stuff, they started saying stuff like, Dad hits mom. And I'm like, whoa, whoa, whoa, guys. They're like, no, put your phone away. Dad, dad and less boys. And I'm like, stop, stop, stop, stop, stop." } ]

[ { "from": "human", "value": "Yeah. What's, what are some secrets, words of wisdom you can say for how to be successful online and all the, everything you've learned. You're kind of like the, the Mr. Beast of comedy, just innovating constantly in terms of how to promote stuff, in terms of how to present stuff across the different social media and across YouTube, podcasts, all that stuff." }, { "from": "gpt", "value": "I think the number one thing is, don't be afraid to suck. No one sees the shit that sucks. Doesn't get views. You just gets forgotten. Yeah, gets forgotten. The good stuff they remember, and then that's all they bring up. The stuff that sucks just goes away, and they're like, oh yeah, did you do that? I don't even think I saw that. Fuck it, who cares? Yeah. You know, it's like, I watched, one of my favorite stories about this is a guy named Sean Patton. I was explaining Instagram stories to him and Ari and Mark Norman. And I was explaining that you can set now sell tickets on Instagram stories. It was a back when they swiped up instead of tap the button to swipe up. And they didn't understand it. And I said, let me show you. And I put up tickets for sale. And I said, you know, hey, guys, swipe up. And then I just showed him in 10 minutes. Look, there's not many people swiped up. There's 145 people. It's a lot of people to buy tickets for a fucking comedy club. And they were like, whoa. And they're like, what do you do stories about? I go, fucking anything. Who gives a shit? Just anything. And some things work, some things don't. And I watched Sean Patton, shout out to Sean Patton, try for about 15 minutes to do an Instagram story about his cup of tea. He was making cup of tea and he was making and he kept going, hey, okay, hold on, hold on. I was filming him from the couch. I was filming all these. The whole thing. Yeah. I filmed every single one of my like 15 different stories of him trying to do that. I had 145,000 views on each story. He never posted it. He never posted it. And I was like, I was like, man, if he had just posted that one, he never posted anything in his stories. That one, 145, 5,000 people would have gone to his story to see that cup of tea being made and see which one he finally went with. And so sometimes I think the biggest mistake you can make is just not doing it. Just do it. Just post it. Turn the camera on. The idea will show up. And that's the number one thing. Just turn the camera on and it'll come up. You'll figure it out." }, { "from": "human", "value": "Do you do multiple takes or no?" }, { "from": "gpt", "value": "Maybe I do three like and I'm talking even like in the dance video I did where I did the hip-hop dance. I did two takes on that With what I had the marching band come to my house. I did two takes on that Once you start doing more takes you lose the fun of it. I think and and then the fun ones I mean my favorite one I've ever done without a doubt my favorite promo I've ever done is I needed they added a second show in some city and, uh, and, and they added a second show like Friday and Saturday and our third show. And I said, I, they just told me, I was like, fuck, I'm leaving like tomorrow. I need to sell tickets. So I said, real quick, girls meet me outside and lands like girls are doing homework. I go to take five seconds. I put on a speedo. I got the American flag. I gave Isla the hose. I gave land the leaf blower and I gave Georgia the drone. And I just had Georgia drone it back to reveal the leaf blower and the mist you were seeing was coming from Isla and Leanne. I'm going to speed it with American flag. I have Bob Seager playing rambling, gremlin man. And that night, Larry, the cable guy texted me and he goes, this is fucking genius. And I was like, it was just, and it was just like, yo, just let's just shoot something. Who gives a fuck? like just shoot something just just just say something and and and put a little just do a little movement into it and and those are my back and that was I used to be I watched a bunch of my promo clips back from back in the day I used to be really good now it's like I'm promoting so fucking many things I'm just like I'm almost like I'm exhausted my promotions yeah because you know the the wall can dry up in terms of ideas Yeah. And it's like, how many times am I going to show a picture of my ass? My ass sells really good. Whatever my ass is on Instagram, fucking million views guaranteed. A million views. Your ass. And it's an active post. Like people are like, I did the best one I did the other day was where I superimposed my ass on my front and it just looked beautiful. No views. Oh, so you got reported. Oh, I see. It looked like a front vagina." }, { "from": "human", "value": "Oh, is that what it was?" }, { "from": "gpt", "value": "Yeah, I think so, but like, but yeah, my ass, like I'm at a place right now where I am promoting. I'll be promoting this movie. I'll be promoting this movie until it goes to a streamer. I'll be promoting it shamelessly. I'm proud of it. I had a great time shooting it. It's in theaters. It'll be on video on demand. It'll be on a streamer. I'm sure I don't know, but I'm sure I'm promoting fully loaded my summer festival with 19 different comics. I'm promoting the tops off world tour That starts again this fall. I'm promoting my cruise the fully loaded cruise. I'm promoting red rocks. I'm promoting so many things That today was the very first time in my career Where I said in January, I'm gonna take a break. I'm gonna take a break and go like Hey man, let's get your resting heart rate down. Let's get some sleep. Let's maybe take off like three months from drinking. Let's just really slow it down and also not think, when you're thinking in promo all the time, it can be a little exhausting. I mean, you think it's exhausting to look at my feet. It's exhausting to be my feet. I'm my feet." } ]

[ { "from": "human", "value": "That's not as less as more. One of the things that, one of the things that inspires me about Rogan, for example, is he almost never does this kind of stuff." }, { "from": "gpt", "value": "Yeah. Yeah. And how suddenly does this podcast? Cause you're right, less is more. What does he do it like once a month? No, he does it three times a fucking week." }, { "from": "human", "value": "More is perfect. But for example, when he does like a story or something like this, he does one take. It's kind of shitty." }, { "from": "gpt", "value": "It's the same. It's the same. It's elk with a bunch of jalapenos, a knife that he had sent to him by a fucking knife maker, and him going, look at this. Look at this, Traeger Smoker. I reverse seared this. This is what's great. Look at this. I'm about to put it in the butter. This is grass-fed butter. This elk really raises my testosterone. Rogan is, you're right. With guys like Rogan, when he does do a story, they're a little precious." }, { "from": "human", "value": "Yeah. And I don't think he's ever done a second take of those stories. No. You have to really admire that. If the shot is framed poorly, if the audio sucks, it doesn't matter." }, { "from": "gpt", "value": "Do you remember when he... This is like, I can get on the weeds on Rogan, and especially when it's Instagram. Do you remember when he got his polar plunge and he couldn't sit it in a minute, and then the next day, and then he must have gotten it on the internet bad. He must have gotten it bad. The next day, he sat in for 20 minutes and he lowered his core temperature by like two degrees and it took him like a month to get that back. That is that fucking man. What an interesting, you know, I told you at the beginning, we were just talking. I like, I really enjoy, I draw inspiration for men I'm not like. You, David Goggins, Cam Haynes. Andrew Huberman, we're talking. Huberman, I love fucking, I mean, there's so many guys like that, Rogan, you know, Tim Kennedy. Like there are a lot of guys that are like just very different men than me and I love Jocko Wilnick. Like I love, I love reading their stuff or listening to their audio books or listening to their podcasts because I'm so not like that. But that took for me when I add a little bit of that in my life. Like I remember you went, I don't know what you were doing, but you were walking us through your day. And it was so fascinating. It was like, it was like, I ran eight miles today. I listened to an audio book. I was running eight miles and then I came and I worked for about three hours straight on AI stuff. And I'm like, and I was like, wait, you sat at a desk? Like you like sat at a desk and like, and I actually, I wouldn't even know what to do. I would start doing crafts. I was like, is your computer?" }, { "from": "human", "value": "Have you set for four hours focused on a single task? Me?" }, { "from": "gpt", "value": "No. Oh, I'm sleeping for four hours. I don't have a brain that... I really admired it when I listened to you do it because then you were like, You were talking, you were fasting at the time, you were doing your fast and then you talk about what you ate. And I was jealous. I wanted to be able to be like that. And I think at that time we started, I think you were doing maybe bone broth or something. And we started doing bone broth pretty religiously and adding that into a fast. Cause I was like, it does for someone like me, it does feel cool to add a little bit of control into life." }, { "from": "human", "value": "So integrating a little bit of a way of being from another person that you're not like," }, { "from": "gpt", "value": "Yeah, like, so David Goggins is consistently pushing himself. I love. I did it over the pandemic more, but I love to go in. Today, I'm running a marathon. I'm going to get on the treadmill, and it's not going to be all at once, but I'm going to get through a marathon today. So let's get up, get up at 6 o'clock for some coffee, and let's see if we can do like five miles, and then know that we started there. And that was really fun. Cam Haines does that too, where he goes like, He almost just sets the limit and then achieves that. His son Truett is really fascinating because his son Truett's right now is trying to get the, and I bet people would never think that this is who I watch, but I get real inspiration from those people because I showed Jaco, I showed a clip to Jaco to Isla because Isla's like me, like a person of her fluids, impulse. I walked in on her one time. She was in the pantry in a bathrobe in the dark drinking a root beer. Yeah. And I opened the door and she went, I said, what? She goes, I thought you were a mom. Shut the door. And I shut the door to let her finish the root beer. Yeah. She has a hard time waking up. So I go, I look, come in here. I want you to see this clip from Jaco. She goes, who's Jaco? I say, he's a Navy SEAL. She goes, he works at the zoo. I went, no. He said, Navy fucking SEAL. I go, it's a badass job. She goes, it's a silly name. And I went, no, it's not. It isn't a silly name. These guys are the toughest dudes in the world." }, { "from": "human", "value": "She goes, she's kind of right though about the silly name. I never even realized that." }, { "from": "gpt", "value": "Well, yeah, you talk to a child. Yeah. And so I go, watch this, what he says about waking up. So it hits the thing and Jaco goes, first thing you do when you hear that alarm clock, you could roll over, you could go back to bed, you could hit snooze, but you get out of bed and that's your first victory. And Isla hits pause and she goes, How great is that nap right after you hit snooze though? It is pretty fun. You're missing the point. He's with Casey Neistat when he's telling the speech. He's with Casey Neistat. And uh, so I hit it and then he goes and then you've got a victory under your belt. Then you go to the gym and you work out and I go, you listen to this island and she goes, yeah. And he goes, and then when you go to the work, now you've got two victories. You've worked out and you've woken up on time. You didn't hit that snooze button. So when you go into that break room and you see Donuts and she hit the pause, she goes, um, if he's about to say no to donuts, I'm done listening. I was definitely going to say no to donuts. Do you think he's going to have a donut? And she goes, You like this guy, I go, I fucking love him. And she goes, do you think he'd like you? And I go, what do you mean? She goes, dad, you hit snooze, you skip working out and you love donuts. And I'm like, yeah, you're making a good fucking point. I go, I, but I, but I do love that. Like I love to watch that brain work and go like, Like I don't, I don't hit snooze. Like I don't hit snooze now. I go, fuck it. Cause I'm up. But I, and then, and then I go to the gym every single morning. I go to the gym every single morning. And today is national donut day. I didn't have a donut. And so like, I, I, like, I try, I like try to apply these a little bit. Cause I, I mean, the other side of me is like, tonight, after we do this, I'll do another podcast and then I'm going to call Leanne and go, yo, we're in Beverly Hills. Meet me out here. Let's go have drinks. Let's go have a fucking couple of compari spritz. I was over at a friend's house and overnight. Let's get some edibles and let's fucking have a night about it. Get some sushi. Go have sex in the hot tub. And I got to work out at fucking 10 in the morning." } ]

[ { "from": "human", "value": "I got to tell you that nap when you press news is the sweetest. That's that's my victory. First victory is pressing snooze and getting the nap onto the belt." }, { "from": "gpt", "value": "Have you ever had, I had, so this morning I didn't work out. I'd press it like five in the morning until about six. You're going hard right now. Yeah, yeah, yeah. And then I had a podcast. I saw you flew to Austin." }, { "from": "human", "value": "I was like, what? Yeah. What? Just run around theaters and now you're back." }, { "from": "gpt", "value": "Yeah, run around theaters to do, to promote the movie. And then look, you only get one shot at these movies. So you might as well do it. And look, it's really hard to get people to go into a movie theater right now. And so I woke up at like six, five or six to seven, did press and then I went And I said, I'm really into a podcast, a history podcast right now. And I was listening to this one about the conquistadors and I was, and I was like, before when I woke up, I was like, I wouldn't mind going back to that. And I said, I'll tell you what, I'm going to take a little nap and go. Now I slept good last night. I slept like seven hours last night. Boy, that 30 minute nap, it felt like five hours. And I woke up going. I gotta readjust my sleep score. That has to be in the 97. That's the best fucking nap. Those little snooze naps are monsters." }, { "from": "human", "value": "You know, there's, I don't care. That's probably for me, one of the peak experiences of life is those naps. I don't know what that is. Have you ever slept in a tour bus? A tour bus versus a regular bus?" }, { "from": "gpt", "value": "Like a tour bus, like meaning like in a bunk." }, { "from": "human", "value": "No. I've taken a lot of buses in my life though. You mean like just like a Greyhound?" }, { "from": "gpt", "value": "Greyhound. Those are very different." }, { "from": "human", "value": "Where people pick your pocket. Listen, Dave. That's just part of life and a Greyhound." }, { "from": "gpt", "value": "I was on the Greyhound one time. I was on the Greyhound one time, right, when Notorious B.I.G.' 's album came out in college. And I was going from Tallahassee down to Tampa. And the dude got on the bus next to me, black dude. And he sat down, he said, Where you headed? And I said Tampa. He said, you grew up in Tampa? I said, yeah. I said, you and he goes, yeah. I said, you go to school with fam. And because of what fam, you and Florida say, the two schools, one's black, one's white. And he goes, no, I just got a prison. I said, you got a prison? And he goes, yeah. Yeah, man. I got out of prison. I stabbed a dude and I've been in prison. I just got out today. He goes, what are you listening to? I said, notorious B.I.G. And he goes, oh, shit. That's out and I went, yeah. And he goes, I want to hear it. Can I hear it? And I was like, yeah. And he used my headset to the whole bus ride to Tampa. I didn't have the, I was like, he's tapped a man. They're yours now. And then I just sat next to him and I was like, is it, are you enjoying it? He's like, this is good. No tour bus snaps." }, { "from": "human", "value": "Yeah." }, { "from": "gpt", "value": "When you're in a bunk. It's ice cold bus isn't moving five o'clock. You got like we work out from like up until like five o'clock five thirty show starts at seven but that five o'clock to six six thirty nap It's pitch black it's like a coffin and it is ice cold and you got all your pillows and blankets in there and you put a little history podcast on in the corner and you just listen to like Kurt Mombashi or like, oh my God." }, { "from": "human", "value": "So you're talking about like five, six PM for a seven PM show." }, { "from": "gpt", "value": "Oh, it's the best nap you'll ever have. You wake up so refreshed. Yeah. Like I've woken up from naps, like where I'm like, fuck yeah. Like those tour bus naps are the best naps." }, { "from": "human", "value": "I think having a full productive day. Mm hmm. And then like 5, 6 p.m. You do a nap and then you do something really stressful after that like really hyper like like a show But if I do like I really like doing podcasts like late at night after a nap" }, { "from": "gpt", "value": "I think I fucked up my podcast rhythm because we're doing them in the day. Yeah. And so sometimes I'm just like, like if I work out the morning, I'm good on a podcast, but if I don't work out, I suck on a podcast. I'm just like, uh-huh. Yeah. Yeah. That's cool. No, I, I'm, my favorite is, I used to do this a lot. I haven't done it in a while is, Full busy day, nap at home, box of wine on the treadmill, and do that until like midnight. Watching TV, box of wine." }, { "from": "human", "value": "Box of wine on a treadmill. It's the best. How are you sipping it, Jake?" }, { "from": "gpt", "value": "You have a glass but you have a box the box fits on the corner in the treadmill So you just got a little spigot. Yeah, and you sit there and just put on like guys grocery games or like just something It's like a you walking or are you running? I walk I walk like a three five So what will happen is if you started three started three casual walk 20 minutes again a mile in and then you and you're gonna feel loose get up to a three five You can do seven miles on these things and just, and just, and you're just enjoying your night. And especially like my wife, back in the day when we lived in our old house, my wife would come into the man cave and she'd sit on the couch and we watched TV together and like watch Game of Thrones. And you just like for an hour and you just get on the treadmill and do it. Oh man, I, I loved having done that as much lately. Cause that was the fucking thing." } ]

[ { "from": "human", "value": "You can lose yourself. You can forget yourself." }, { "from": "gpt", "value": "You forget you're on a treadmill and you just start like walking and you're sweating and you're, and you're having wine and then certain points you're like, Like, uh, you, like, you, you start jogging, especially like we were watching, uh, not all quiet on the rest in front. What's the one where they've got to, they've got to get the message to the front line. I would run like I was in World War two. I would run when they would run and it was fucking awesome. And I would be like, cause they were doing drinks. They'd be like, take a drink. You got to run up there and be like, okay. Yeah. I'm really into, I'm really into history podcasts right now. Like I've been in them for a while. 20th century or everything? Everything. Everything. I listened to, uh, uh, was it Marcus Aurelius and Cleopatra today? Uh, in the middle of the podcast, they're like, you know, just, you know, Cleopatra wasn't as hot as everyone says. I was like, why are you saying that? Yeah. Tell me she's a six. She looked like Snooki. Yeah. Like, let me, let me dream that she looked like Kim Kardashian. Yeah. And then I listened to that one today." }, { "from": "human", "value": "I listened to World War II a lot. Oh. A lot of books on World War II, a lot of podcasts about it." }, { "from": "gpt", "value": "What's your favorite theater, the Pacific or European?" }, { "from": "human", "value": "No, European. Pacific, well, because my family's so much of it has died in that theater. Dude, so much of it. I hate to sound cliche, like I'm Arnold Schwarzenegger in an 80s movie, but it's personal." }, { "from": "gpt", "value": "Well, you know what I find funny about that? Yeah. Is Lenin, It was Lennon, right? John Lennon? No, no, no. Lennon was World War II, right? Or was it Stalin? Stalin. Stalin thought him and Hitler would be friends. Yeah. I think that's so cute that he was like, we're going to be buddies. Yeah. Like we're boys. Yeah. And then Hitler's like, oh, fuck you. I'm Hitler, bro. This is what I do." }, { "from": "human", "value": "Well, at first he was like, fuck yeah. Yeah. Yeah, yeah, yeah. We'll be friends." }, { "from": "gpt", "value": "Yeah, yeah, sure, sure. And then call me, call me. Better yet, I'll show up and win her. How's that sound? Yeah. That, uh, that, uh, I find that I find all of that so fascinating. When I was a kid, I had a hard time understanding how, cause we were always at war with Russia when I was a kid and I had a hard time understanding how we could be on the same team as Russia in World War two. And I was like, wait, how did that turn south? And then you, this is a podcast. Yeah. I'm, I'm that all of that, all of that, like, All of the World War II, I got really into the Pacific shit for a while, but I'm really deep in the, in the whole, we went to Normandy from the Cliffs of Dover and to think, so that's the fucking beach. That's the fucking beach. Oh shit, this is cold as fuck. These dudes jumped in ice cold water. And this isn't like a cool beach, like this is, And it's just so insane to, and then to see like the, it's pretty intense. I don't know why I got into history late in life. I wasn't, I was kind of into it, but I'm really into it now. And I love just learning a little something about, you know, today was, I listened to like the Noiser has a great history stuff. It's like history for dummies, I think. Like I get, I can't listen to the really smart stuff. It starts losing me. Too many names that I'm gone." }, { "from": "human", "value": "Yeah. Well, Dan Carlin is always the best." }, { "from": "gpt", "value": "Your podcast with Dan Carlin was so fucking good. He is. He is. I mean, I only know about Martin Luther because of Dan Carlin. I only know. Like I would have, I made a joke about some Lutheran the other day. And everyone laughed and I was like, did you guys listen to the same podcast? I did because I just am quoting Dan Carlin. I that his podcast, I mean, we all quoted the same shit about Genghis Khan about how he could shoot a dove off a horseback. But we knew that all four hoofs had to be off the ground at the same time they shot. They would sometimes take their meat and stick it under their saddles. And that, I mean, Dan Carlin is He is the motherfucker. Yeah. God, those were ma'am discovering podcasts was the coolest time of my life." }, { "from": "human", "value": "And the interesting thing about him is, is audio. And so it's the voice. So for me, I mean, everybody has a story with a prostitute in a motel. Dan Carlin was my prostitute in the motel. I invited him to a motel. I don't even remember in the middle of nowhere. He shows up with that voice and the rest, the rest is history. Things they will never know what happened. Oh boy, did he like, no, I'm just kidding. I'm really running with it. Yes, we did we got along with it. I mean, he's an incredibly kind and thoughtful person constantly self-critical thinking he's not good enough He just works on those episodes forever forever and he just thinks that there's these just like with Goggins. There's these other historians That I think like way on him like they're going to judge him or something like this in his mind I talked about I talked to it with them, but Daniel Patelli and I guess" }, { "from": "gpt", "value": "His thing is, I guess his hang-up is, Carlin's is, he's not like this, he doesn't feel like he's a real historian, but he is. I mean, it's the way he shares the story. Look, we had a history teacher in seventh grade who was a con artist, and he came and taught her seventh grade, and it was a total con artist. All he taught us about was barnstorming, barnstorming and the JFK assassination. I walked away having such an appreciation for barnstorming, a very small blip on Americana. But I know a lot about barnstorming because of this guy and he, because he was passionate about barnstorming. It was the funniest seventh grade history class I've ever taken my entire life." } ]

[ { "from": "human", "value": "Uh, you mentioned tour bus. Uh, what's the best road trip you've been on? Have you, have you driven across the country?" }, { "from": "gpt", "value": "Just you know, we had one road trip in college. where a bar had burned down and we found out that the alcohol was still in the bar. So we took my Jetta over to the bar and we emptied out the bar of all the burnt up liquor. We didn't know what anything was. We threw it into the trunk with no clothes, like no clothes. And we said, let's just start driving. And we drove from Tallahassee and ended up in West Virginia and just drove. Just drove. It was like five dudes. We drank. One person drove. I wasn't a big drinker. And this sounds kind of crazy. I wasn't a big drinker in college until I went to Russia. And so I drove the majority of the way and then we get somewhere. and we'd get a hotel room. And it was one of the most epic fucking weekends of just debauchery and chaos. And I think we took mushrooms and we went down a river in tubes." }, { "from": "human", "value": "So it's mostly about the bond between the five people versus like throwing yourself into Meeting like strangers and this kind of stuff." }, { "from": "gpt", "value": "Yeah, and it was when it was when I was like I was trying to be a poet at that time and so like we were actual poetry Yeah, it's really bad part of my life. I was just we were like is like music is there too somewhere Yeah, I tried to be in a band too that didn't work out the way I wanted it to I just was never I always was I was never the I'm not a good serious person Like I'm a good funny person. I'm good to be the fool, but I'm not good when I try to be serious It looks foolish And so even like, you know, someone said, showed me throwing out the first pitch the other day as something and all I could see is my gut flying everywhere. And I was like, that's why it went viral. I was like, fuck, I thought it was my throwing style. That was a pretty epic one. The best road trips I ever took was during the pandemic. During the pandemic when we were, when we would, we were doing that drive-in movie theater tour. I created a drive-in movie theater tour when no one was working. And the very first one, no one had left our houses. We all went to my house. We got my tour bus. The day the tour bus showed up, we just wrapped the cabin. Just wrapped doing the cabin on Netflix and we got COVID tested We got in the tour bus and we said let's create our bubble and stay safe and that night we drove I think just outside Gallop and we stayed in a No one had been outside and we stayed at a KOA camp on arrival and we watched thunderstorms come in on a lake and we were all smoking cigars and drinking IPAs and it was fucking mad and that whole road trip. We rode tripped across the country to start the tours. And then we just and we were outside for the first time when we were and it was that period of my life. I'm so grateful. I didn't make a ton of money because we had a big crew and I made sure I was trying to make sure everyone because no one had paid bills in a while. But that fucking tour was the funnest. We took up pickleball. We took up disc golf. Everything was wide open in the middle of the country. You could do things, but we were still in our bubble. And we lived in that tour bus. And at night, we just get back in the tour bus by ourselves and get fucking wasted. And I mean, that fucking run, it'll nothing will ever be like that." }, { "from": "human", "value": "Yeah, because you like rediscovered the humanity, the camaraderie, because the pandemic kind of killed that. It suffocated us from like their basic connection." }, { "from": "gpt", "value": "Yeah. And there was this You always had an anxiety attack halfway through that you got COVID. This one, you know, COVID was killing people and you were like, I know I've got COVID. And then, and then we'd come in to Sedona. You know, this was also why I didn't make a ton of money. We'd come into Sedona and we'd get a house in Sedona for five days at Airbnb. And we'd all isolate there before we went home to our families. I remember the first time we did it, we didn't isolate. We just pulled back into LA and all our families were freaked out. So I had to stay but live on the tour bus at a different house. And it was 4th of July and I came in and they were doing an egg toss. And they were like, well, you can do an egg toss. This is not about my anxiety guys. You can do an egg toss with your daughter Isla. So Isla and I did an egg toss and we fucking got it to like legit like 40 fucking feet where I was throwing overhand lobbing them to her. And the egg cracked in her hands and it broke over her face and I got in bed that night high on weed. fucking drunk and I went, what if I gave my daughter COVID from an egg toss? I go, that would be the most horrific way to kill a child. Yeah, she got it from an egg toss. It broke in her face. Her dad did COVID." } ]

[ { "from": "human", "value": "Yeah. I mean, that's the scariest thing about COVID, I think, is not you dying, but because of you, somebody else dies. Oh, I got it. That's the fear, I think, that really Like took hold on people. That's why they're they were way more cautious than perhaps they needed to be just like I don't want to be the guy that's responsible for killing somebody I love or somebody that somebody else loves But Leanne gave everyone in our family covered She came up to me and she gave me a hug." }, { "from": "gpt", "value": "Yeah after Christmas And gives me a hug and I said what's the matter she goes I think I have covid I go bitch yeah fuck you doing yeah back it up Fauci like fucking and then she we went when we went all. And then and then she had it and it went away and we're all fine and we're like okay what we still go to park city so we all go to park city then georgia gets covered. And then Georgia starts crying, and she goes, I think I gave you COVID. And I was like, I'm fine. She goes, you're at high risk. And I'm like, what? She's like, you're a fat alcoholic. I'm like, whoa, easy. Where the fuck? What about feelings? And then Georgia got COVID. Me and I live, we're still in Park City. Leanne had it. My tour bus came up, picked George up. George and Leanne took him back to LA. And I was, I don't know that we're there. And I tested and then we get on the plane and I cough and Isla's got her mask on and Isla just looks at me, you know, the best. The one thing we miss are the eyes of her masks. And then we get home and I had COVID and Isla's like, if you gave me fucking COVID and then she never got COVID, she never got COVID." }, { "from": "human", "value": "You mentioned IPAs. What's the lay of the land of great beers in this expansive world of ours?" }, { "from": "gpt", "value": "I like Pilsners. I'm a Pilsner guy. I'm an ice cold beer guy." }, { "from": "human", "value": "Ice cold Pilsner." }, { "from": "gpt", "value": "I like an ice cold beer. I like an ice cold. One of the best beer buzzes I ever had is we had bought a new house, but we weren't building it. And over COVID, I didn't drink for like three months, four months. until we decided to go back out on tour and we figured that out. And I was in the backyard and Leanne came back with like two tall boy German Pilsners and cracked them. She goes, this type of beer. And I was like, fuck yeah. And man, that second that beer buzz is different than anything. Cause the second it, it touches you, you feel the sparkle and you're just like, Yeah, baby. I'm back. And then she's like, you want another one? I'm like, fuck yeah. And then we have another one. And then, and that beer buzz is especially afternoon beer buzz is just so pretty. It's different than a night beer buzz. A night beer buzz is like, oh, I guess we'll have a beer and then I'm a couple. I gotta go to work tomorrow. All right. But that afternoon, irresponsible beer buzz. Yeah. Pilsner. Pilsner, there's, I don't mind an IPA, but it's gotta be at, it's gotta, I gotta be at somewhere. I gotta be at somewhere. They gotta have an IP, like I like a local IPA. It's gotta be food involved. And then I'm enjoying an IPA. I threw up an IPA in a bathtub in Salt Lake City one time. It was bad." }, { "from": "human", "value": "What do you mean? Oh, you threw up in the bathtub. Okay." }, { "from": "gpt", "value": "No, no, no. I was drinking in the bathtub." }, { "from": "human", "value": "Like bathtub with some water in it or? No, no, no, no, no. You're sitting in the bathtub?" }, { "from": "gpt", "value": "No, I'm getting ready to take a shower. And I had a growler of IPA. Yeah. It's about six in the morning. Yeah. I have to fly that day to Vegas to jump off the stratosphere. Yeah. All I need is a little bit of a buzz to get myself on this fucking plane. Yeah. I have anxiety going through my fucking head because I'm gonna be the first person to jump off the stratosphere. And so I'm drinking an IPA out of a growler in a bathtub at six in the morning. And it just didn't sit right. And I threw it up and I was off IPAs for like 10 fucking years. I didn't have another IPA until I was like 46, because I was like, all I could think about was throwing up that IPA in a bathtub in a shower and being like, I'm gonna throw up thinking about it. But yeah, I don't, I didn't fuck with IPAs for a while. And then during the pandemic, I got it, I got back into IPAs." }, { "from": "human", "value": "I was that way with Jack and Coke, Jack Daniels and Coke. Really? Got so drunk many, many years ago that just couldn't look at it for a few years." }, { "from": "gpt", "value": "I can't touch. What's Janice Joplin's drink? Southern Comfort. I can't say Southern Comfort." }, { "from": "human", "value": "She just drank it straight? Well, Southern Comfort, is that 40%? That's hard." }, { "from": "gpt", "value": "No, it's almost like a liqueur. Honestly, it's really sweet. Jack, double Jack on the rocks, lots of rocks. I had a quote that, I got a video that went viral about, I'll never quit drinking. And I have not gotten on a plane since that without having a flight attendant not even ask what I want. Walk up with a double jack on the rocks, lots of rocks. I mean, it happens. So on times I don't want to drink. Yeah. I flew to Austin the other day. I was like, I'm not drinking on the plane because I felt sick. Yeah. And man, he walked up and he goes double jack on the rocks, lots of rocks. I was like, fuck, you know it. Yeah. But yeah, that's, that's my drink. Uh, uh, wait, what's your drink? I thought you'd be a vodka guy." } ]

[ { "from": "human", "value": "Well, vodka I associate with a lot of moments in my life of happiness because vodka is associated with shots and camaraderie. And there's Eastern Europe in general. A bunch of guys get together. and you celebrate life intensely. People fight, people just feel, people experience life in a deep way, just get too drunk for like for a reason. But I don't see that as like drinking, drinking. That's like a journey towards a destination. To me, drinking is like whiskey neat. or beer, like you said, ice cold beer. This is that afternoon beer, like for a person like me who's extremely regimented and disciplined, sometimes you get the situation. I just remember, I mean, just, I remember there's a bar at MIT where I just remember in the afternoon, there was like a business meeting and I sat down, we just started with a beer and they were like really crisply cold. Some kind of lager and then just one beer and two beers and all of a sudden you say fuck it Let's just the rest of the day. This is what this is and you said bright like you just like and nature is beautiful out there and you just said fuck it to all the meetings fuck it to everything else just the camaraderie, the just talking about life, being silly, being, you know, all of that." }, { "from": "gpt", "value": "See that, see, you're describing something I'll never get, because I'm not regimented. I would love, I would love. I would love only for the moment that you get to have, you get the most precious little angels breath that anyone's gonna get because you live this life that's dialed in and it's scheduled. And so when you say, fuck it, it's like, that's like me doing coke. Like it's like, oh, fuck it. And you can just do that with beer. That is fucking amazing. I would love to live a regimented life just to get the chance to go Hey guys, I'm not performing surgery today. Fuck it. Hey guys, this bus full of kids, kids gotta drive itself home. Fuck it. I couldn't even, I couldn't even tell you. You know what's so interesting about me and you is like, you know, I'm a fan of yours. I couldn't even, I don't really know what AI is like other than like dildos or AI, right? Like," }, { "from": "human", "value": "Like if I had some electronics in it. No, I don't think a dildo classifies this a robot." }, { "from": "gpt", "value": "Cause I thought that's how they were going to take over fucking mankind. Well, they start with these, the fucking womanizer that knows how to eat a pussy. And then they're like, we don't need men anymore. But like, it's so interesting that I couldn't even, like I can barely tell you what my dad does for a living. He's a lawyer. I could kind of, like he's called me and say, Hey man, you got to stop talking about what I do. Cause you're not doing it right. I go, what do you mean? He's like, whatever you're saying I do, I don't do. And so, but like, I couldn't even begin to explain your job to my daughter." }, { "from": "human", "value": "Well, I think there's the way you do the job, which is the way you experience life that I've regimented, just even just a 95 job, right?" }, { "from": "gpt", "value": "Do you work 95?" }, { "from": "human", "value": "No, but I work much more than that. Yeah. But there's no. Every kind of lifestyle has its complexities. I think nine to five means you have to wake up at a certain time every day." }, { "from": "gpt", "value": "Do people tell you to pull it back and then like, hey, you're working too hard?" }, { "from": "human", "value": "Yeah, a lot of people. What I want in my life, what I love in my life, especially people close to me are the ones that say work harder. Because everybody- How do you get them? How do you get them? I don't have anyone in my life that does that. Everybody's always worried about me." }, { "from": "gpt", "value": "Everyone's worried about me. Everyone's always worried about me. The only thing, no one ever says, if I showed you my DMs right now on my texts, all it is is like, hey man, I'm worried about you. Like Joe, I love Joe to death. Not once has he, he always, I don't know if you've seen the caring Joe eyes where like I'm worried about you, where it's like, And you're like, man, don't do that. And he's like, I'm worried about you. And you're like, don't fucking worry about me. I'm working. I'm just working. I love working. I love what I do. I love what I do. I can only imagine that you love what you do as much as I love what I do, because that's all I ever want to think about. It's all I ever want to talk about. It's all I ever want to be like, when my wife wants me to take August and whatever the August whatever the fuck September that's no, but like July and August off after we do fully loaded and I Was like no I was like what do you mean? Like we're so like well just what am I like sitting a chair and stuff? No, you can do like take up a hobby. I go writing jokes Can I get on stage? Can I do sets in the city because I'm not gonna fucking you? Do you want me to like like I don't understand what I don't understand people who don't I don't understand people don't don't have the drive to work and all the time, I don't understand that. I envy it. I envy it because I go, I would love to just be like, I'm going to play video games. I can't play a video game. If I did, I'll be on Twitch trying to make money. Like I love how do you find somebody who goes, you need to work harder." } ]

[ { "from": "human", "value": "Yeah. Especially when they see like, I don't know what you look like, but I look tormented when I'm working very hard. Like it means Like it's easy to worry for my well-being. It's not like I'm happy go lucky, but I'm happy underneath it." }, { "from": "gpt", "value": "My working too hard is like on a plane every other day or every day getting up at six working out and then going into like two in the morning or one in the morning and then going to sleep and then getting up at six working out getting on a plane going to another place. Like that's what my working is just a lot of travel in mind. If I didn't drink on planes, I don't think anyone would probably worry about me at all. They'd be like, no, you're doing great." }, { "from": "human", "value": "But the working out is there. So I got to ask you, because you and I seem to have a similar relationship with running. I like this idea of slow treadmill. I like running slow. Yeah. You know, like a 10 minute mile or something like this. I'll just run forever. Just listen to an audio book, listen to brown noise. Think. I love it. What's brown noise? It's actually brownian noise, but short brown noise. It's kind of like white noise, but deeper. So it's just noise like that. Oh, wow. It kind of sounds like a waterfall. And I'll just listen to it and something happens to me. I think there's, you can ask Kuberman about this. I don't, but apparently there's some science that kind of focuses the mind and so on." }, { "from": "gpt", "value": "It's funny because I think sometimes people don't get about running as they go, They get on and they put it on like a six right away on like a treadmill and they go, I can't do this for fucking 30 minutes. But you can if you build yourself up to your six and you play games. I play games with myself a lot. Like I'll play games where I go. Um, and I can do this. I can do this. I can do this almost not all day long, but I can do this a lot longer. Like I did it for two hours the other day before my workout or no hour and a half before my workout where I started at 3.5 walking. And then I, and then I get my jog up to like a five and then I'm like 5.5. Then I go six. And then I go, and once I'm at a six, my body's really loose. And so then I go, all right, bring it down to a 3.5. And then I go punch it to a 7.5 for a quarter of a mile. And then let's walk at a four, let's jog at a five. I can do that. I love, they're called a fartlets, I think. And I love those like where it's like, it's, Regiment and running and sprinting at certain levels. I love that and I love also when I work out to listen to music but put on foreign language cooking shows on Netflix Can you explain that they just put it on mute? You don't need to hear it for some reason. They're sexier the way they're shot. They're sexier. Yeah, and it's like Street food Asia, and they're just watching them make street food in Asia. There's one called broth. It's called broth the nation of broth and they're just making broths and you're like Fuck I want some broth." }, { "from": "human", "value": "So this is South America. This is Asia South America Asia" }, { "from": "gpt", "value": "They do one on pizzas. I've watched this one on pizzas. I've watched it. There's like four episodes. I've watched every episode so many times that my trainer goes, God, that's that fucking, I could watch that again. I love watching foreign language cooking shows on mute when I work out." }, { "from": "human", "value": "While blasting music. While blasting hip hop. See Darren. All right. Well, I usually run outside, but I do the same thing." }, { "from": "gpt", "value": "Oh, I can't. I don't mind hiking. And I like, I do, like, you know what, my fun thing, the fun thing I would do, and this is a, this is ripped off totally from Cam Haines, is I like to go, there's a great hike called Fryman Canyon out in the valley where I live. And I'd be like, all right, I'm going to run to Fryman Canyon. I'm going to jog to Fryman Canyon. I'm going to hike Fryman Canyon. And then I'm going to jog back. And then I wonder how long that is. don't think about it. Just know that that's your thing. Just know that that's your thing. And then all of a sudden you're like, shit, that was eight miles. You're like, that's a fucking beast. The hike doesn't even feel like anything. The hike, once you do the jog, there's a hard part there. The jog home is a fucking cakewalk. I had the best jog when I was in Serbia. I had the best jog. I think it was in Hyde Park. I think the name of the park's Hyde Park. It was, get this, it was 80% downhill. Wow. So this jog, You started the park and it was like, and I'll do it like this for you. You went like, down hell like this, down hell like this, down hell like this. And then it was like a little steep incline to get to the beginning of it. So it was like a little like nice, like lower back pain. You're like, okay, okay. And then the jog was perfect. And you did it four times. It was four miles. I loved it. My very first day, I did it. A butterfly got in front of me. It's like six in the morning. So yeah, six in the morning. The butterfly got in front of me. And it ran the whole first mile in front of me. And I just, it was like so surreal. It would disappear and then it would show back up in front of me. You sure it was real? Definitely real. And then the next day, and so I go, I'm on the right path. I'm on the right path. The next day I go for a jog, six in the morning, and it rained. And as I'm doing the steep uphill incline, a tree collapsed in front of me. Just went, and I went, whoa, a tree fell in the woods and I was here to see it. Shut the fuck up. I was like, that's crazy and it didn't hit me. I'm on the right path and then I started getting overthinking it like the next day I saw condom and I was like, this is a good place to do it. I was like, people fucked here last night. But, but yeah, that was one of the best jogs I've ever had. I could do that. I could run that park every single day and no one jogged that park in Serbia." } ]

[ { "from": "human", "value": "I actually recently had condom on my doorstep and I wanted to ask whoever is up there above what that means, what the You know, what the story behind that is. Yeah." }, { "from": "gpt", "value": "Like, you know, just a pissed off UPS guy. I was it compromised? What do you mean compromised? Like, was there stuff inside it?" }, { "from": "human", "value": "Oh, I didn't look." }, { "from": "gpt", "value": "I mean, very few people put on the condom and go, it's not for me. I'm going to take it off. Yeah. It's a little tight." }, { "from": "human", "value": "It could be to send a message like a horse's head kind of thing. Oh, yeah. I don't know. I don't know what the message is exactly." }, { "from": "gpt", "value": "Do you use condoms? If you had one next time, would you use a condom?" }, { "from": "human", "value": "I don't know what a condom is. I've learned about sex from you today." }, { "from": "gpt", "value": "If my wife dies, there's no condom to my future. I'll get it. I don't care. I have babies. I don't give a shit." }, { "from": "human", "value": "You're a man of principle. I like it. Mark Marin wrote something about your book many, many, many ages ago, Life for the Party. I think there's an interesting question in there. So first of all, I wrote eloquently about you. Burke Christcher is one of the great American wild men, a Gonzo warrior driven not by cynicism or desire to reveal a dark truth, but instead by a deep, almost essential need to have a good time, no matter what. His stories track the trials and tribulations of a big hearted dude trying to fit in, help out and party and find himself. After all is said and done, we arrive with him at the true humility of joy. So when was this desire for pure joy born in you? It seems like you are driven towards this joy." }, { "from": "gpt", "value": "Yeah. I think I think I arrived at it in college. I wasn't always the guy, I was really a serious, pretty serious dude, like when I got to college." }, { "from": "human", "value": "Like dark brooding serious?" }, { "from": "gpt", "value": "Yeah, I wanted to be like Eddie Vedder, I wanted to be like a poet and a fucking lead singer and I wanted to be taken seriously and I wanted to be attractive and I wanted to be, you know, like, I wanted to be wanted by chicks and respected by dudes and, you know, And slowly but surely, I think the bird I really am chipped away at that and was like, Hey man, like I would say funny. I would say funny things that would surprise me. They were so funny. Meaning like in my fraternity, I'd say things that were the funniest thing that I go. How did I think of that? And then slowly I was like, and I remember as that, as I chipped away at that and as I got older, I would I would, I would lean towards that. We would go on a bus trip to like Clemson and I would get on the, the, the walkie-talkie on the bus and I would do stand up for like an hour. I just make jokes for an hour. And I, and I loved the impulse of like, you know, you get on the bus and everyone's like, we're going to drink and you get that one group of girls that wasn't going to drink and then to break them. And they're like, oh, fuck it. Fine. Let's drink. And then that to, to watch that happen. And I think, Then once I, and then when I went to Russia, this is so cliche for me to say, but interacting with those young banditi, they were, they were pretty serious dudes. They didn't have like, they were pretty stoic dudes. And when you could make them laugh, it was like a real, it was a real joy, the silliest things. I remember, you know, we were told, There was a club, it was called Cafe Europa or maybe Cafe Americana. There's where everyone really hung out, like all the real dudes, the real dudes in St. Petersburg. And they had told us that like, they told us in our class that Russian women didn't have tampons. So they were like, bring tampons to give to the Babushka, the Davushka that ran the floor." }, { "from": "human", "value": "After all this time, the fact that you're Russian sucks is awesome. Because you keep mispronouncing words horribly." }, { "from": "gpt", "value": "Horribly. I'm so fucking bad. I am so bad. You know that's how I got the name of the machine, right? Yeah. I was trying to say I'm a man. And I said I'm a car. And they just were like, huh? And so I came into the room one night with a top pocket full of tampons. And they were like, what are those for? I said, girls, this is how we're going to pick up girls. And the look on their face was so, it was so pure joy. I remember the first night I pulled out lemons. And this is, it sounds, it sounds make believe now to say it because it's been in the movie. And if I brought out lemons, we were going to do lemon drops. I had vodka and lemons and sugar and a pocket knife and a fanny pack. And I put out lemons and this guy Igor goes, oh, the machine runs on lemons. Like it was just so foreign that you would need lemon for them. And so I think, and I made friends with those guys. Like, I mean, like friends, whatever you can, but like, definitely me and that guy Igor were legit friends, legit friends. Like," }, { "from": "human", "value": "but from a place of joy. Yeah, it was like real seeing them light up." }, { "from": "gpt", "value": "And then I remember backpacking through Europe after that and realizing I could bring a spark of fun to like a campfire. And I had little tricks like I learned all the currencies around the world. And so I challenge people to a currency game. I go, let's go head to head. You say, you name a country in the currency and I'll name a country in the currency. And I would, I would slow roll them. I just do all the dollars, right? The Chinese dollar, the American dollar, the Australian dollar. I'd run through all the dollars and they'd be like, you only know the dollars. And then I, and then I get, start getting really deep. I could make an event out of a night in Europe. I could make an event. I remember we were in, uh, Strausberg and the bar was supposed to close, but I had friends coming to the bar. They had the train hadn't gotten, uh, in, in yet. And I said, I said, don't close the bar. I'll stand on the bar and I'll entertain everyone in this room. I'll dance. I'll do whatever if you just keep the bar open. And the guy goes, if you can keep me laughing, I'll keep the bar open. And I danced and, and made this guy laugh on a bar for legit. I'm not even like legit 30 minutes until my friends walked in and they're like, what the fuck's going on? I had this ability to. It's without humility. I had mine looking like a fool and like doing and there were really stupid jokes at the time just very like base American frat boy jokes But I had a spark when I came back I feel I feel like I learned how to like To do that and do funny things that were like fun like I would we go to a football game and I'd I'd I brought a camera around with me a lot I like to take pictures And I remember one of the jokes I'd go is I go, hey, everybody, I told my parents I have a lot of friends. Can you guys gather around and we'll take a picture? And so like 50 people would lean around like, hey. And like just the dumbest, silliest things. And I wasn't a big self promoter. So these would happen. And I was just doing them for me and my friends. I'd bring a guitar out and I'd just make up songs. and and so I think in in watching this this serious bird get chipped away at until my like six and a half year of college where it was just this silly guy and then Rolling Stones written about me and I'm like and then I'm like well and I remember a couple serious dads going you know you got a chance to turn this into something you need to go to need you need to follow your dream and I tried to stand up the first night and I was like Maybe not wasn't the funniest guy, but man, they laughed and we had a good time for 30 minutes and it worked and it was stream of consciousness and I was like, yeah, and then of course, you know that same little path you take gets convoluted in in New York because you want to be a serious comic and you want to be taken seriously and you want to be edgy and you want to say the and then slowly but surely that you're the real Bert chips away at it and he's like, yeah, man Just fucking take your shirt off, kill a beer. Like fucking tell the machine story. Tell the story about your stupid kids. Like just, you don't need to be the fucking edgiest dude in the room. Leave that to the good ones. Let Bill and fucking Joe and Tom, they're the, those are the legit fucking gangster comics. They're, I mean, those are the fucking You're never gonna out fucking Chappelle those those guys are fucking gangsters do what you do have fun Make life fun and seize the day and fucking try to bring a spark to people and let them forget about their fucking day and the the interesting thing about alcohol that I don't really like drinking In terms of the actual physical like how it makes me feel but there's the that the camaraderie that happens what I" } ]

[ { "from": "human", "value": "I wish there was another way to get there, but I don't think there is. That joy you get of just everybody getting together and taking shots or drinking, not the polite kind of drinking, but just everybody just, the fuck it kind of drinking." }, { "from": "gpt", "value": "Yeah. And when someone goes, all right, I'm in. Like that's the funnest. And then especially when you have a big group and there's like a big group and like we were at the premiere the other night and I was working, I was legit working. I'm trying to sell the movie to outlets, media outlets. And bust them with the boys, my wife flew them in to surprise me. And they're over on the red carpet. And they're like, yo, let's scull a beer. Let's kill a beer. And I was like, oh, and then I'm like, oh, fuck it. I don't give a fuck. This is what life's about. And that moment where they're like, yeah, that's the funnest fucking, especially when I can get Leanne to do it. When I get Leanne to get a drink and she's like, Like I was supposed to drink last night and she was like, you want to just. We still have a glass of wine in the front yard and you're like, oh, my heart skipped a beat. I go glass wine in the front yard. Are we going to talk wild? Are we going to say crazy shit to each other? Like, fuck yeah. In the front yard. Front yard. What are you doing in the front yard? Let the dogs play. Oh, big trees in the front yard. Why are you sitting on the, where were you sitting? Sitting in the Adirondacks on the front porch, letting the dogs play, letting the dogs rest in the front yard. Maybe go over, pick some cherries, look at the fucking trees, hear the wind going through the front yard, and then you go to the backyard. That's where the cigar comes out, a little more privacy, but yeah. Red wine? Red wine, maw back. Keep our soul fights low." }, { "from": "human", "value": "Yeah wine is a whole another one too. I feel sophisticated. Oh, yeah, I've I've went I traveled before Ukraine I went to Paris for the first time and drank wine there and I felt like I Was a sophisticated man, you know dude I mean to speak with French, but you don't need to you don't need to just start smoking to just have a fucking cigarette I was gonna start writing poetry to the Hemingway." }, { "from": "gpt", "value": "There's like a glass of wine in Italy and like like just out in like we went we went ah the most beautiful fucking day of my life Just outside Florence They have all the vineyards and stuff, and we took a Vespa tour. Nice. And it was fucking, and Ila's too young to ride, so she's on the back. And now everyone can drink there, you know? You're not supposed to drink and drive Vespas, but a glass of wine is nothing. And you're like, okay, I guess I'll have a couple then. If you think one's nothing, then I think two's nothing also." }, { "from": "human", "value": "So this is drunk Bert on a Vespa?" }, { "from": "gpt", "value": "drunk, not even drunk, but just lightly, lightly feathered where, and it's beautiful. My wife hasn't been drinking, Georgia has, Georgia's too young, Isla is too young, but I had, at dinner we had a couple and we're going through these hills and it starts raining and it starts raining and I go, Isla, what are we missing? And she goes, music, big boy. And I go, what do you want to hear? And then she just starts going, uka, shaka, uka, uka, uka, shaka, uka, uka. I can't stop this feeling. She's screaming it behind. And then I hear Georgia screaming it. And then Leanne singing it. And then Sandy and her two daughters are singing it. We're all in these vespas. And we're all singing it. And I swear to God, if I hadn't been lightly feathered, Don't know if I would have been like I know what are we missing cuz you know alcohol will give you that thing like what's next? What's bigger? How do we take this to the next level? And then I got this little girl who's my still my little girl at that moment when she's singing sing it all the good times And then we stop at the light and all of us are looking and it's pouring rain. It is, I mean, I remember sobbing crying, sobbing crying, sobbing crying. And then I was like, let's get these fucking Vespas up and get to a vineyard. And then we went to a vineyard and all the girls tried a little wine. No one got buzzed or anything, but you're just sitting there with your family going like, Fuck, dude." }, { "from": "human", "value": "And I'm not certain you get that without a little bit of... A little bit, but then a lot is also interesting. I had a lot of whiskey with Rogan once." }, { "from": "gpt", "value": "I think I saw that. Yeah. I think I saw that with Whitney too. And the push-up concert contest." }, { "from": "human", "value": "The only time I met David Goggins in person, one of the great embarrassments of my life is trashed out of my mind. I'm trying to just say how much of a fan I am. And I could tell by the look in his eye that what is actually coming out of my mouth is not good. It's not good. And his wife is embarrassed standing right there also. Just everybody's embarrassed for me. Man, I'm trying my best. And then somehow I find myself a minute later doing pushups with him as a challenge. Pushups? I didn't, it wasn't counting. All I knew I'm not stopping. And he was polite enough to stop. I think probably like 50, I don't know." }, { "from": "gpt", "value": "There's a gift someone gives you when they get drunk and they allow, they get drunk and they lower their guard and they allow you to see them that way. It's a gift. It's a real gift. It's just as cool as like pulling someone aside and going, yo, I'm into anime. Like are you cool with that? And then they're going, yeah, I'm, oh, yeah. Can I show you some cool anime? And then you're like, that's not my thing, but it's like a cool thing. It's like sharing cool music with someone or like telling them something about your childhood. When you, when you get drunk, like real drunk around someone, You're giving them a gift. And that gift is, I want you to see me for everything I am. You're not gonna get any lies. You're not gonna get any bullshit. This is me at my absolute worst. I have a rule. I will never, never get mad at someone for getting too fucked up. because they're giving you a gift. They're saying like, yo, I did this on accident, but I trust you enough to let you know that this is who I am, and this is what you're getting. I love it. I love it. Especially when it comes to someone you don't expect. Like you? Shut the fuck up. David Gawking should have hugged you and said, thanks for showing me this." } ]

[ { "from": "human", "value": "Yeah, I guess that really haven't you're right. I haven't thought of it that way is there's an intimacy to that is that is a gift of like fragility like this is me. I mean, I'm a very loving drunk, but I'm boy am I a dumb drunk." }, { "from": "gpt", "value": "I'm just I bet you haven't seen dumb drunk." }, { "from": "human", "value": "Well, dumb in terms of the eloquence of the words that are coming out of my mouth. So like, you're just not stitching words together correctly to make sentences. No. I turn into Hemingway lost in translation. Like, I'll start saying random words. but like translated to Japanese and back to English. It's not making any sense, but there's love. Like I think what you feel is this kind of desire to connect that I think I always feel towards other people, but when I'm drunk, let me just say, fuck it. I don't need to be polite. I'll just be loving fully." }, { "from": "gpt", "value": "Yeah. Yeah. It's, uh, it's cool when people let you see that in them because some people are really like measured. There's not one drink, and then I'm done. I don't want it, no, no, no, no. And you're like, oh, show me a little more. And then like, it's cool. And I never, like, I love, my favorite moments in the next morning when they're like, did I make an asset of myself? And you get to say, not at all. Not at all. That's the coolest thing. Because some people, bad people, will leverage that over you. And you can see it's a bully tactic. They'll leverage that, the bad interactions I've had in this business that we were talking about earlier. They were people that, When you got fucked up, they used it against you. They held it over your head and they said, listen, man, I don't know what the fuck, but if you want, and you're like, hold on. And then, and then they made you, they gaslighted you into thinking you were the only one that was fucked up. And you're like, okay, I guess I know who you are now. And I was like, didn't you get fucked up a little bit too? Like, I know we smoked weed and I thought we ate Xanax, but like, what about the, I know I had a couple beers, but like, I wasn't that fucked up." }, { "from": "human", "value": "Well, Whitney's really good at that because she's the person I woke up next morning and she just said, you're wonderful, you're eloquent, you're great. I'm sure she was making that up completely." }, { "from": "gpt", "value": "It was funny. I got blackout drunk at Whitney's roast of me, like blackout drunk. Don't remember any of it. And don't remember any of it. Don't remember speaking on the microphone. I don't remember talking in the microphone. And I did the set apparently. Yeah. And I said to Whitney, like in a real moment of vulnerability, Whitney's like a sister to me, like a legit sister. My daughters call her Aunt Whitney, Leanne, she's like a sister to Leanne. If she comes over to do a podcast, Whitney will show up an hour early and sit with Leanne out in Leanne's Rose Garden and just talk. And Whitney, I said the next day I called Whitney and I was getting on a plane and I knew I was hungover. I said, how bad was I last night? She goes, you were fucking fantastic. and just didn't let me sit with it. And then I watched the video and I was like, oh, I was fucking out of control. I was way out of control. Thank you for lying to me." }, { "from": "human", "value": "So while being a wild out there comic, she's also this caring and loving human being." }, { "from": "gpt", "value": "So interesting. She has the most nurturing sensibility. She is, she sends my daughter's gifts out of nowhere, just out of nowhere to send Georgia gifts up at college. And just a gift box, like she is just, she takes care of people." }, { "from": "human", "value": "That's her love language. And she's also the person that taught me what the hell love language even means. Apparently there's a book and there's like five of them." }, { "from": "gpt", "value": "Oh, you and Whitney, I would love to sit in the fucking room and watch you two talk." }, { "from": "human", "value": "Yeah, she has a robot that looks like her. And that's the future, that robot." }, { "from": "gpt", "value": "I told her she should have that robot dupe porn. Let someone fuck that robot. And then she gets deep faked to face if she wants, but I go, promote it. It's promoted special that way." }, { "from": "human", "value": "Like a full-on, not like showtime, softcore porn, like full-on porn." }, { "from": "gpt", "value": "Full-on, like the dude like, he's like, hi, welcome inside. You look tired, do you need a drink? Oops, I'm stuck in the dryer, you know." }, { "from": "human", "value": "What's the promotion you're most proud of that you've done?" }, { "from": "gpt", "value": "Without a doubt, it's gotta be when I took dance lessons. Yeah, that was epic. Because I tried to get Joe and Tom and Ari to do it for sober October. I was rooting for you. And they said, no, Joe's like, absolutely not. You knew Joe was gonna say no. But Joe's a legit dancer. Like he can legit dance. He took like dance lessons for a movie and is apparently an amazing dancer." }, { "from": "human", "value": "And so... You're right. He could have said yes. He could have said yes, because I remember listening to that podcast where you really made the pitch." }, { "from": "gpt", "value": "Yeah. And I said, we all take hip hop dance lessons. It's your, your cardio is through the roof and in those, you do an hour dance lesson. And then we all do a hip hop dance. And then we all put our hip top dances against each other. And we see who has the best hip hop dance. It's different than the typical stuff we do. It's fun. And we can also work out, we can do everything, but like, let's do hip hop dance. And Joe Adam Lee was like, that's a horrible idea. I will not do that. And then Tom and Ari, just whatever, they were like, yeah, I side with Joe, we'll do something else. And then I was like, but I knew I liked hip hop dance. And so I was like, I'm gonna take hip hop dance lessons anyway. I'm gonna just do it over sober October and I'm just gonna, I'm gonna fucking do it. And then maybe I'll do a hip hop dance video and then show it to them and let them rate me. And then we did it and it was $1,200 for hip hop dance lessons for like a month or whatever for like a couple weeks. And then when we did it, I thought, you know what, I bet we could put tour dates next to this. Cause it's, it was, it turned out being like kind of good, but kind of funny. I thought we could put tour dates next to this. And it was the beginning of me figuring out the whole like, my promos were short and quick. And they were always like something silly. What's up guys? It's your boy Burke, right? Sure. And then that one blew up and immediately, I was announcing the tour and immediately every tour date sold out. And then every, we added shows that day we added shows and every one of those sold out. And I was like, whoa. And then Tom was like, the fuck was that? And I was like, I think I just sold out my whole fucking tour. And, and then Tom, and then when Tom did his one like that and put his tour dates on it went over the top. He sold out his tours and and he was like whoa, he's like do I sold out every fucking show we've added shows I'm doing like an arena here and this is before we were doing arena So I was like shut the fuck up and that changed my game where I was like all right So everything's gotta be and then and then I was like I don't mind spending a little money and then the next one I did was the marching band And I was just like and but it was you know, it's real crazy. I just said Can you guys get me a marching band? And they were like, yeah, it's like 2400 bucks. I go, cool. And they're like, what do you want them to do? And I go, I don't know. We'll figure it out. And they came over and they had outfits. And they were like, so what do you want us to play? I go, what song do you guys know? And then they didn't even know each other. They were all hired people. And they were like, rubber band man. And I was like, oh, sure. Yeah. And then we just scripted it out. I was like, all right, I'll pretend it's a regular thing. Let's don't wait till I walk out and then I'll blow a whistle. And then you come out with the dum, dum, dum, dum, dum, dum, dum. And then, and then we got like two videos out of that and that sold out dates and then I was like, fuck. And then, and then, and then, and then, and then the real baller one, the fucking real one, the real one was I tore all the muscles in my arm going into, into shooting the movie and I ripped all the tricep muscles off and they were retracted. So I had to go in for surgery and the anesthesiologist said to me, the night before, he's like, what music do you want going in, going under? And I was like, what? And he goes, you can pick your music going under. I said, wait, hold on. What is it? Like, I go, is it like casual? And he was like, yeah. I said, I'm doing a promo read going under. He goes, what do you mean? I said, I'll explain it to you tomorrow, but I'm going to, I'm not going to count backwards. I'm just going to do a promo read until I pass out and my phone's going to fall on the ground. All I need you to do is hit stop record and. So they rolled me into surgery. I had red rocks. I had only sold like 75% of red rocks, which is the biggest, 10,000, the biggest venue I'd ever played in. And I rolled in with my thing with CCR playing in the background. And I just was like, I was like, I lay here on this mat or on this steel pile. I was terrified as I go into surgery. I want you to know one thing. I'm playing Red Rock September 10th, and I started doing my promo read and he was like, yeah, I go Jimmy Buffett's there the night before and the night after. He's like, for real? I go, yeah, you want to come? He's like, hell, yeah. He's like, all right. He's like, you're going under. And he hits it. I go, sure. And I dropped the phone. I woke up out of surgery. And the first thing I said is, where's my phone? And he goes, you got it. And Leigh-Anne's like, I've already watched it. It's good. And I went, for real? And so that was a big one. That was a big one, too." } ]

[ { "from": "human", "value": "Yeah, that was genius, but that was sort of opportunity, kind of jumping in an idea. Have you ever met Mr. Beast, Jimmy? No. So he's, I got to hang out with him for a day and listen to Brainstorm with the team ideas. He kind of sounds like you, but on steroids in terms of like, first of all, willing to spend any amount of money on an idea. Like anything, anything is allowed. Anything is allowed. I love that energy. The only thing is no like rated R stuff. So he's really trying to go like for the broader audience for kids and so on. So that's the limitation. So you can't do like most of the ideas that pop into my head when anything is allowed is kind of Dark yeah, but the point is to really aggressively brainstorm every single day at the Y board like what and he does that for YouTube shorts too for one minute videos Really and just like really like what is the coolest thing we could possibly do? I love that. I love that energy. And that energy, most people don't do that and they should. Basically, YouTubers do that because they're obsessed about this particular YouTube algorithm and so on. And Jimmy is the best at that. But like, that can benefit you if you're a scientist, if you're a comedian, if you're whatever." }, { "from": "gpt", "value": "Just go all out. There's something really authentically fun about Just coming up with a really stupid idea and just going let's see what happens who gives a fuck we've had those like I thought I could catch an arrow one time and Yeah, and my wife's like absolutely fucking not and I go I really think I can I go let's just get really close shoot it And i'll see if i can catch it and she's like no that's not what you're doing and then my cousin's like hey we can cheat it to make you look like you caught an arrow and i went what he was just okay let's do it like this and so we did it they got sold a ton of tickets it looked like i really caught it everyone knows i didn't catch an arrow i mean for the movie we did a promo legendary was like you do all the promos and we did a promo and i thought i could slide out of my car like tom cruise did In a in a helicopter and with roller skates and get behind the car and then skate behind the car and we just ran over my foot and it was real and it was scary. but we just caught her all, and in doing it, my cousin's like, give me the read. And I'm like, duh, my movie comes out Memorial Day weekend. I think we have to go to the hospital, and then we went to the hospital, and I'm in the Speedo with a helmet on, one broken roller skate, and then that did really well. And so the whole thing is just turning the camera on. You never know what's gonna fucking happen." }, { "from": "human", "value": "And don't be afraid to look stupid and all that kind of stuff, just go all out." }, { "from": "gpt", "value": "Fucking ego's the death of comedy. Like, when you really give, like, it's hard to be the coolest guy in the room and still want to be a comedian. You just let yourself be whoever you are. And you'll see the great ones, they're like that. They don't, you know, they're just regular fucking dudes. And you get some real slick ones. Like, Dave's probably the coolest guy in the fucking room. He really might be. But I think that's just who the fuck he is." }, { "from": "human", "value": "Who are the, outside your close friend group, who are the greats to your top five?" }, { "from": "gpt", "value": "outside of my friend group?" }, { "from": "human", "value": "So, like for me, probably Norm MacDonald, Mitch Heidberg, George Carlin, Tom Segour, number one. Oh, yeah? No." }, { "from": "gpt", "value": "I wanted to say that. It's an interesting pairing. No, Mitch Heidberg, David Tell, David Tell." }, { "from": "human", "value": "David Tell. Everybody deeply respects David Tell. More than even his comedy, I love the show where he was... Insomniac. Insomniac. He's so good at just the natural comedy of human interaction." }, { "from": "gpt", "value": "He's a brilliant comic. He's just a brilliant fucking mind. Norm McDonald's a fucking genius, an absolute genius. I mean, look, you know, Chapelle and Bill Burr are the two best in our generation, in my opinion. And that's, you know, and that's understanding that, like, I'm still friends with some of the best in our generation, but, like, just the way their brains work is really on the next level. Like, you know, those are guys. Stan Hope is the same way. Just fucking genius. You know, Rogan, Seguro, those are my friend group. That's my really close friend group. But, like, those other guys I'm friends with, but they're not like that. Yeah." }, { "from": "human", "value": "Louis C.K." }, { "from": "gpt", "value": "Oh, Louis C.K. Fuck. He's amazing. It's hard to do because you start like forgetting the people and then it's almost like people go, oh, so you didn't like that person? You're like, no, I fucking love that person." }, { "from": "human", "value": "I mean, there's, for me, like Robin Williams, which is a whole other thing." }, { "from": "gpt", "value": "Oh, yeah. Sarah Silverman. Fucking, I love Janine. I'm a fucking big Janine fan. I love her brain. Marin's fucking hilarious. You know, Patton's brilliant. David Cross. I mean, there's like, it's, you know, it's really to ignore those guys. And those guys were like the whole forefront of the alternative comedy front. They're fucking amazing." } ]

[ { "from": "human", "value": "Yeah. And the people that had like Marin or Joe that have podcasts is interesting. It's an interesting pairing because you get to know like everything about them, but then they're also comedians. And I've gotten, I've gone to Joe's club a lot. Listen, Joe and just that whole group of comics do comedy and go to the same set over and over and over to see how it changes. It's really cool." }, { "from": "gpt", "value": "It's the coolest thing when I first started to watch someone like a tell was really brilliant at it because you'd watch him tinker with a set. Yeah. And he'd have an idea like hitting someone over the head with a hammer from behind. And then he'd work it nine different ways, you know, and you'd be like, whoa. So the hitting hammer in the head is the thing that he's working on." }, { "from": "human", "value": "I hope Eddie Murphy comes back. He's one of the greats." }, { "from": "gpt", "value": "Chris Rock. Chris Rock's fucking bring the pain is like the reason I got into comedy. Yeah." }, { "from": "human", "value": "Do you have advice for young folks? You seem like a stellar example of successful human being. Or more seriously, a very kind of nonlinear life. Do you have advice for young people in high school and college? How to have a good life they can be proud of?" }, { "from": "gpt", "value": "Don't worry too much about what you do when you get older. But when you do start worrying, find the thing you love and it'll never feel like work. To fucking, I can't imagine what it must feel like to be a lawyer and have to read papers or whatever they do all day and know that that's my day. I mean, it must feel like what school felt like for me. If you find that thing you love to do, You will work endlessly, effortlessly, and hardest fucking shit every day, and you'll love every day of your life." }, { "from": "human", "value": "Fine. Would you love and let it kill you? Bacowski said that." }, { "from": "gpt", "value": "Then that's the fucking thing. Cause that's what I'm doing. I'm letting it fucking take. I'm, I'm fucking, I'm, I'm on that fucking bullet train to fucking nowhere on this fucking comedy journey. I love it. I love it." }, { "from": "human", "value": "So it's like the Hunter S. Thompson thing, however you put it, basically you're not supposed to end up in a well-preserved box. You slide it in sideways, smoke, just beat up, just a giant mess." }, { "from": "gpt", "value": "I love those guys that can really live that life that are like, yeah, man, I'm breaking it. I'm breaking it down. Like a guy like Tony Hawk or Matt Hoffman, who's like, yeah, if I don't fucking break the fuck out of this thing before, then I'm not doing it right." }, { "from": "human", "value": "And then again, there's also Churchill who tried to break the thing and he couldn't. Lived into his 90s, a bad motherfucker. Just a bad motherfucker. So I guess the only hope we can have for Burr-Crisher is that you're gonna be the Winston Churchill of comedy." }, { "from": "gpt", "value": "My life has lived perfectly. If I'm 95 years old and all my friends are dead. Yeah. And they're like, can you believe that wolf got Rogan? Yeah. Can you believe, Segura, 600 pounds, and they had to carry him out in a crane out of that house. Can you believe that it's just Bert, Chrysler, and Joey Diaz left?" }, { "from": "human", "value": "Just standing there in October. So, in October, just the two of you left. Just laughing." }, { "from": "gpt", "value": "In October, it's just I'm 95 at the Mark Twain Awards, and they're like, Bert, Chrysler, still fucking here. Who saw that coming?" }, { "from": "human", "value": "I hope so. Bert, I'm a huge fan of yours. It meant so much to me when you said that you knew who I was like we were talking about. It's just, it's so cool. It's so cool. Thank you for putting so much joy and love out there in the world. I'm a huge fan and thank you for the love you give me." }, { "from": "gpt", "value": "Thank you, man. I am a huge fan of yours. You keep doing you. I'll keep doing me. And then let's see if we can meet in the middle sometime. Have a cold beer in the afternoon." }, { "from": "human", "value": "Love you, brother." }, { "from": "gpt", "value": "Love you. Thanks for listening to this conversation with Bert Kreischer. To support this podcast, please check out our sponsors in the description. And now let me leave you with some words from Hunter S. Thompson. Life should not be a journey to the grave with the intention of arriving safely in a pretty and well-preserved body, but rather to skid in broadside in a cloud of smoke, thoroughly used up, totally worn out and loudly proclaiming, wow, what a ride. Thank you for listening. I hope to see you next time." } ]

[ { "from": "human", "value": "Please check out our sponsors in the description and now with your friends, here's Chris Lattner. It's been, I think, two years since we last talked and in that time, you somehow went and co-created a new programming language called Mojo. So it's optimized for AI, it's a super set of Python. Let's look at the big picture." }, { "from": "gpt", "value": "What is the vision for Mojo? For Mojo, well, so I mean, I think you have to zoom out. So I've been working on a lot of related technologies for many, many years. So I've worked on LLVM and a lot of things in mobile and servers and things like this, but the world's changing. And what's happened with AI is we have new GPUs and new machine learning accelerators and other ASICs and things like that that make AI go real fast. At Google, I worked on TPUs. That's one of the biggest, larger scale deployed systems that exist for AI. And really what you see is if you look across all of the things that are happening in the industry, there's this new compute platform coming. And it's not just about CPUs or GPUs or TPUs or NPUs or IPUs or whatever, all the PUs, right? It's about how do we program these things, right? And so for software folks like us, right, it doesn't do us any good if there's this amazing hardware that we can't use. And one of the things you find out really quick is that having the theoretical capability of programming something and then having the world's power and the innovation of all the smart people in the world get unleashed on something can be quite different. And so really where Mojo came from was starting from a problem of we need to be able to take machine learning, take the infrastructure underneath it and make it way more accessible, way more usable, way more understandable by normal people and researchers and other folks that are not themselves like experts in GPUs and things like this. And then through that journey, we realized, hey, we need syntax for this." }, { "from": "human", "value": "We need to do program language, right? So one of the main features of the language, I say so fully in jest is that it allows you to have the file extension to be an emoji or the fire emoji, which is one of the first emojis used as a file extension I've ever seen in my life. And then you ask yourself the question, why in the 21st century are we not using Unicode for file extensions? This means it's an epic decision. I think clearly the most important decision made in most, but you could also just use Mojo as the file extension." }, { "from": "gpt", "value": "Well, so, okay, so take a step back. I mean, come on, Lex. Do you think that the world's ready for this? This is a big moment in the world, right?" }, { "from": "human", "value": "This is, we're releasing this onto the world. This is innovation. I mean, it really is kind of brilliant. Emojis are such a big part of our daily lives." }, { "from": "gpt", "value": "Why is it not in programming? This is innovation. Well, and like you take a step back and look at what file extensions are, right? They're basically metadata, right? And so why are we spending all the screen space on them and all this stuff? Also, you know, you have them stacked up next to text files and PDF files and whatever else. Like if you're gonna do something cool, you want it to stand out, right? And emojis are colorful." }, { "from": "human", "value": "They're visual, they're beautiful, right? What's been the response so far from, is there support on like Windows on the operating system in displaying like file explorer?" }, { "from": "gpt", "value": "Yeah, yeah, yeah, yeah, yeah. The one problem I've seen is that Git doesn't escape it right. And so it thinks that the fire emoji is unprintable and so it like prints out weird hex things if you use the command line Git tool. But everything else, as far as I'm aware, works fine. And I have faith that Git can be improved. So I'm not sure. And so GitHub is fine. GitHub is fine, yeah. GitHub is fine, Visual Studio Code, Windows, like all this stuff, totally ready because people have internationalization in their normal part of their path. So this is just like taking the next step, right? Somewhere between, oh wow, that makes sense. Cool, I like new things too. Oh my God, you're killing my baby. Like what are you talking about? This can never be, like I can never handle this. How am I gonna type this, like all these things. And so this is something where I think that the world will get there. We don't have to bet the whole farm on this. I think we can provide both paths," }, { "from": "human", "value": "but I think it'll be great. And so GitHub, when can we have emojis" }, { "from": "gpt", "value": "as part of the code, I wonder? Yeah, so I mean, lots of languages provide that. So I think that we have partial support for that. It's probably not fully done yet, but yeah, you can do that. For example, in Swift, you can do that for sure. So an example we gave at Apple was the dog cow. So that's a classical Mac heritage thing. And so you use the dog and the cow emoji together, and that could be your variable name. But of course the internet went and made pile of poop for everything. So if you wanna name your function pile of poop, then you can totally go to town and see how that gets through code review." }, { "from": "human", "value": "Okay, so let me just ask a bunch of random questions. So is Mojo primarily designed for AIs, or is it a general purpose programming?" }, { "from": "gpt", "value": "Yeah, good question, yeah, good question. So it's AI first. And so AI is driving a lot of the requirements. And so Modular is building and designing and driving Mojo forward. And it's not because it's an interesting project theoretically to build, it's because we need it. And so at Modular, we're really tackling the AI infrastructure landscape and the big problems in AI. And the reasons that it is so difficult to use and scale and adopt and deploy and all these big problems in AI. And so we're coming at it from that perspective. Now, when you do that, when you start tackling these problems, you realize that the solution to these problems isn't actually an AI specific solution. And so while we're doing this, we're building Mojo to be a fully general programming language. And that means that you can obviously tackle GPUs and CPUs and these AI things, but it's also a really great way to build NumPy and other things like that. Or just if you look at what many Python libraries are today, often they're a layer of Python for the API and they end up being C and C++ code underneath them. That's very true in AI, that's true in lots of other domains as well. And so anytime you see this pattern, that's an opportunity for Mojo to help simplify the world" } ]

[ { "from": "human", "value": "and help people have one thing. To optimize through simplification by having one thing. So you mentioned Modular. Mojo is the programming language," }, { "from": "gpt", "value": "Modular is the whole software stack. So just over a year ago, we started this company called Modular. Okay, what Modular is about is, it's about taking AI and up leveling it into the next generation, right? And so if you take a step back, what's gone on in the last five, six, seven, eight years is that we've had things like TensorFlow and PyTorch and these other systems come in, you've used them, you know this. And what's happened is these things have grown like crazy and they get tons of users. It's in production deployment scenarios, it's being used to power so many systems. I mean, AI is all around us now. It used to be controversial years ago, but now it's a thing. But the challenge with these systems is that they haven't always been thought out with current demands in mind. And so you think about it, where were LLMs eight years ago? Well, they didn't exist, right? AI has changed so much and a lot of what people are doing today are very different than when these systems were built. And meanwhile, the hardware side of this has gotten into a huge mess. There's tons of new chips and accelerators and every big company's announcing a new chip every day, it feels like. And so between that, you have like moving system on one side, moving system on the other side and it just turns into this gigantic mess, which makes it very difficult for people to actually use AI, particularly in production deployment scenarios. And it's what modular is doing is we're helping build out that software stack to help solve some of those problems so then people can be more productive and get more AI research into production. Now, what Mojo does is it's a really, really, really important piece of that. And so that is part of that engine and part of the technology" }, { "from": "human", "value": "that allows us to solve these problems. So Mojo is a programming language that allows you to do a higher level programming, the low level programming, they do all kinds of programming in that spectrum" }, { "from": "gpt", "value": "that gets you closer and closer to the hardware. So take a step back." }, { "from": "human", "value": "So Lex, what do you love about Python? Oh boy, where do I begin?" }, { "from": "gpt", "value": "What is love, what do I love about Python? Your guy who knows love, I know this." }, { "from": "human", "value": "Yes, yes. How intuitive it is, how it feels like I'm writing natural language English, how when I can not just write, but read other people's code, somehow I can understand it faster. It's more condensed than other languages, like ones I'm really familiar with like C++ and C. There's a bunch of sexy little features. Yeah. We'll probably talk about some of them," }, { "from": "gpt", "value": "but list comprehensions and stuff like this. Yeah. Also, and don't forget the entire ecosystem of all the packages. Oh yeah, there's probably huge. Because there's always something," }, { "from": "human", "value": "if you wanna do anything, there's always a package. Yeah, so it's not just the ecosystem of the packages and the ecosystem of the humans that do it. That's a really, that's an interesting dynamic. That's good. I think something about the usability and the ecosystem makes the thing viral," }, { "from": "gpt", "value": "it grows and then it's a virtuous cycle, I think. Well, and there's many things that went into that. So I think that ML was very good for Python. And so I think that TensorFlow and PyTorch and these systems embracing Python really took and helped Python grow. But I think that the major thing underlying it is that Python's like the universal connector, right? It really helps bring together lots of different systems so you can compose them and build out larger systems without having to understand how it works." }, { "from": "human", "value": "But then what is the problem with Python? Well, I guess you could say several things," }, { "from": "gpt", "value": "but probably that it's slow. I think that's usually what people complain about, right? And so slow, I mean, other people complain about tabs and spaces versus curly braces or whatever, but I mean, those people are just wrong because it is actually just better to use indentation." }, { "from": "human", "value": "Wow, strong words. So actually, I'm a small tangent. Let's actually take that, let's take all kinds of tangents. Oh, come on, Lex, you can push me on. I can take it. Design, listen, I've recently left Emacs for VS Code. Okay. The kind of hate mail I had to receive because on the way to doing that, I also said I've considered Vim. Yep. And chose not to and went with VS Code. You're touching on deep religions, right? Anyway, tabs is an interesting design decision and so you've really written a new programming language here. Yes, it is a super set of Python, but you can make a bunch of different interesting decisions here. Totally, yeah. And you chose actually to stick with Python as in terms of some of the syntax." }, { "from": "gpt", "value": "Yeah, well, so let me explain why, right? So I mean, you can explain this in many rational ways. I think that the indentation is beautiful, but that's not a rational explanation, right? But I can defend it rationally, right? So first of all, Python 1 has millions of programmers. Yeah. It is huge, it's everywhere, it owns machine learning, right? And so factually, it is the thing, right? Second of all, if you look at it, C code, C++ code, Java, whatever, Swift, curly brace languages, also run through formatting tools and get indented. And so if they're not indented correctly, first of all, we'll twist your brain around, it can lead to bugs. There's notorious bugs that have happened across time where the indentation was wrong or misleading and it wasn't formatted, right? And so it turned into an issue, right? And so what ends up happening in modern large scale code bases is people run automatic formatters. So now what you end up with is indentation and curly braces. Well, if you're gonna have the notion of grouping, why not have one thing, right? And get rid of all the clutter and have a more beautiful thing, right? Also you look at many of these languages, it's like, okay, well you can have curly braces or you can omit them if there's one statement or you just like enter this entire world of complicated design space that objectively you don't need" } ]

[ { "from": "human", "value": "if you have Python style indentation, so. Yeah, I would love to actually see statistics on errors made because of indentation. Like how many errors are made in Python versus in C++ that have to do with basic formatting, all that kind of stuff." }, { "from": "gpt", "value": "I would love to see. I think it's probably pretty minor because once you get, like you use VS code, I do too. So if you get VS code set up, it does the indentation for you generally, right? And so you don't, you know, it's actually really nice to not have to fight it. And then what you can see is the editor is telling you how your code will work by indenting it," }, { "from": "human", "value": "which I think is pretty cool. I honestly don't think I've ever, I don't remember having an error in Python" }, { "from": "gpt", "value": "because I indented stuff wrong. So, I mean, I think that there's, again, this is a religious thing. And so I can joke about it and I love to kind of, you know, I realize that this is such a polarizing thing and everybody wants to argue about it. And so I like poking at the bear a little bit, right? But frankly, right, come back to the first point, Python one, like it's huge, it's an AI, it's the right thing. For us, like we see Mojo as being an incredible part of the Python ecosystem. We're not looking to break Python or change it or quote unquote fix it. We love Python for what it is. Our view is that Python is just not done yet. And so if you look at, you know, you mentioned Python being slow. Well, there's a couple of different things that go into that, which we can talk about if you want. But one of them is that it just doesn't have those features that you would use to do C-like programming. And so if you say, okay, well, I'm forced out of Python into C for certain use cases, well, then what we're doing is we're saying, okay, well, why is that? Can we just add those features that are missing from Python back up to Mojo? And then you can have everything that's great about Python, all the things you're talking about that you love, plus not be forced out of it when you do something a little bit more computationally intense" }, { "from": "human", "value": "or weird or hardware-y or whatever it is that you're doing. Well, a million questions I want to ask, but high level again, is it compiled or is it an interpreted language?" }, { "from": "gpt", "value": "So Python is just in time compilation, what's Mojo? So Mojo is a complicated answer, it does all the things. So it's interpreted, it's shit compiled, and it's statically compiled. And so this is for a variety of reasons. So one of the things that makes Python beautiful is that it's very dynamic. And because it's dynamic, one of the things they added is that it has this powerful metaprogramming feature. And so if you look at something like PyTorch or TensorFlow or I mean, even a simple use case, like you'd find a class that has the plus method, right? You can overload the dunder methods like dunder add, for example, and then the plus method works on your class. And so it has very nice and very expressive dynamic metaprogramming features. In Mojo, we want all those features come in. Like we don't want to break Python, we want all to work. But the problem is, is you can't run those super dynamic features on an embedded processor or on a GPU, right? Or if you could, you probably don't want to just because of the performance. And so we entered this question of saying, okay, how do you get the power of this dynamic metaprogramming into a language that has to be super efficient in specific cases? And so what we did was we said, okay, we'll take that interpreter, Python has an interpreter in it. Right? Take that interpreter and allow it to run at compile time. And so now what you get is you get compile time metaprogramming. And so this is super interesting and super powerful because one of the big advantages you get is you get Python style, expressive APIs. you get the ability to have overloaded operators. If you look at what happens inside of PyTorch, for example, with automatic differentiation and eager mode and all these things, they're using these really dynamic and powerful features at runtime. But we can take those features and lift them, so that they run at compile time." }, { "from": "human", "value": "So your C++ is metaprogramming with templates," }, { "from": "gpt", "value": "but it's really messy. It's super messy. It was accidentally, I mean, different people have different interpretations. My interpretation is that it was made accidentally powerful. It was not designed to be Turing complete, for example, but that was discovered along the way accidentally. So there have been a number of languages in the space, and so they usually have templates or code instantiation, code copying features of various sorts. Some more modern languages or some more newer languages, let's say, they're fairly unknown, like Zig, for example, says, okay, well, let's take all of those types that you can run it, all those things you can do at runtime, and allow them to happen at compile time. So one of the problems with C++, I mean, which is one of the problems with C++. Here we go. Wrong words. We're going to send everybody to it. That's okay. I mean, everybody hates me for a variety of reasons anyways, I'm sure. I've written enough. Just the way they show love. I've written enough C++ code to earn a little bit of grumpiness with C++. But one of the problems with it is that the meta programming system templates is just a completely different universe from the normal runtime programming world. So if you do a meta programming and programming, it's just like a different universe. Different syntax, different concepts, different stuff going on. So again, one of our goals with Mojo's to make things really easy to use, easy to learn." } ]

[ { "from": "human", "value": "So there's a natural stepping stone, for example. You're going to send everybody to it." }, { "from": "gpt", "value": "I freaking. As you do this, you say, okay, well, I have to do programming at runtime, after you're programming at compile time, why are these different things?" }, { "from": "human", "value": "How hard is that to pull it out, because that sounds, to me, as a fan of meta programming in C++, even, how hard is it to pull that off? That sounds really, really exciting, because you can do the same style programming at compile time and at runtime." }, { "from": "gpt", "value": "That's really, really exciting. Yep, yep. And so, in terms of the compiler implementation details, it's hard. I won't be shy about that, it's super hard. It requires what Mojo has underneath the covers is a completely new approach to the design of the compiler itself. And so this builds on these technologies like MLIR that you mentioned, but also includes other caching and other interpreters and JIT compilers, and other stuff. Show you a duplicate of the inside the compiler. Within the compiler, yes." }, { "from": "human", "value": "Show you a duplicate of the interpreter." }, { "from": "gpt", "value": "Within in the compiler, yeah. And so it really takes the standard model of programming languages and kind of twists it and unifies it the runtime model, which I think is really cool. To me, the value of that is that, again, many of these languages have metaprogramming features. They grow macros or something. Lisp, right? I know your roots, right? This is a powerful thing. If you go back to Lisp, one of the most powerful things about it is that it said that the metaprogramming and the programming are the same. That made it way simpler, way more consistent, way easier to understand, reason about, and it made it more composable. If you build a library, you can use it both at runtime and compile time, which is pretty" }, { "from": "human", "value": "cool. Cool. For machine learning, I think metaprogramming, I think we could generally say, is extremely useful. I'll jump around, but the feature of autotuning and adaptive compilation just blows my mind." }, { "from": "gpt", "value": "Yeah. Let's come back to that. All right. What is machine learning? What is a machine learning model? You take a PyTorch model off the internet, right? It's really interesting to me because what PyTorch and what TensorFlow and all these frameworks are pushing compute into is they're pushing into this abstract specification of a compute problem, which then gets mapped in a whole bunch of different ways. This is why it became a metaprogramming problem. You want to be able to say, cool, I have this neural net. Now run it with batch size 1,000. Do a mapping across batch. Or okay, I want to take this problem, now run it across 1,000 GPUs. This problem of describe the compute and then map it and do things and transform it, actually it's very profound, and that's one of the things that makes machine learning systems really special." }, { "from": "human", "value": "Maybe keep checking. All right. Can you describe autotuning and how do you pull off... I mean, I guess adaptive compilation is what we're talking about as metaprogramming. How do you pull off autotuning? Is that as profound as I think it is? It seems like a really... We'll mention list comprehensions. To me, from a quick glance at Mojo, which by the way, I have to absolutely dive in. As I realize how amazing this is, I absolutely must dive in." }, { "from": "gpt", "value": "That looks like just an incredible feature for machine learning people. Yeah. Well, so what is autotuning? Take a step back. Autotuning is a feature in Mojo. Very little of what we're doing is actually research. Many of these ideas have existed in other systems and other places, and so what we're doing is we're pulling together good ideas, remixing them, and making them into hopefully a beautiful system. So autotuning, the observation is that it turns out hardware systems, algorithms are really complicated. It turns out maybe you don't actually want to know how the hardware works. A lot of people don't. There are lots of really smart hardware people, I know a lot of them, where they know everything about, okay, the cache size is this, and the number of registers is that, and if you use this length of vector, it's going to be super efficient because it maps directly onto what it can do. All this kind of stuff. The GPU has SMs, and it has a warp size of whatever. All the stuff that goes into these things, or the tile size of a TPU is 128, these factoids. My belief is that most normal people, and I love hardware people also, I'm not trying to fen literally everybody on the internet, but most programmers actually don't want to know this stuff. So if you come at it from perspective of how do we allow people to build both more abstracted but also more portable code, because it could be that the vector length changes, or the cache size changes, or it could be that the tile size of your matrix changes, or the number in A100 versus an H100 versus a Volta versus a whatever GPU have different characteristics. A lot of the algorithms that you run are actually the same, but the parameters, these magic numbers you have to fill in end up being really fiddly numbers that an expert has to go figure out. So what autotuning does, it says, okay, well, guess what? There's a lot of compute out there. So instead of having humans go randomly try all the things, or do a grid search, or go search some complicated multi-dimensional space, how about we have computers do that? And so what autotuning does is you can say, hey, here's my algorithm. If it's a matrix operation or something like that, you can say, okay, I'm going to carve it up into blocks, I'm going to do those blocks in parallel, and I want this with 128 things that I'm running on, I want to cut it this way or that way or whatever." } ]

[ { "from": "human", "value": "And you can say, hey, go see which one's actually empirically better on this system." }, { "from": "gpt", "value": "And then the result of that, you cash for that system, you save it and save it. And so come back to twisting your compiler brain, right? So not only does the compiler have an interpreter that you use to do programming, the compiler that interpreter that metaprogramming now has to actually take your code and go run it on a target machine, see which one of likes the best and then stitch it in, and then keep" }, { "from": "human", "value": "going. Part of the compilation is machine specific." }, { "from": "gpt", "value": "Yeah. Well, so I mean, this is an optional feature, right? So you don't have to use it for everything. But yeah, if you're so one, one of one of the things that we're in the quest of is ultimate performance. Yes. Right. Ultimate performance is important for a couple of reasons, right? So if you're an enterprise, you're looking to save costs and compute and things like this, ultimate performance translates to fewer servers. Like if you care about the environment, hey, better performance leads to more efficiency, right? I mean, you could joke and say like, you know, Python's bad for the environment. Right. And so if you move to Mojo, it's like at least 10 X better just out of the box and keep going. Right. But but performance is also interesting because it leads to better products. And so in the space of machine learning, if you reduce the latency of a model so that it runs faster, so every time you query the server running the model, it takes less time, well, then the product team can go and make the model bigger. Well, that's actually makes it so you have a better experience as a customer. And so a lot of people care about that." }, { "from": "human", "value": "Yeah. Hey, you know. So for auto tuning for like tile size, you mentioned 120 for TPU. You would specify like a bunch of options to try just in the code. Just simple statement. And then you just set and forget and know, depending" }, { "from": "gpt", "value": "what wherever it compiles, it'll actually be the fastest. And yeah, exactly. And the beauty of this is that it helps you in a whole bunch of different ways. Right. So if you're building so often what will happen is that, you know, you've written a bunch of software yourself, right? You you wake up one day, you say, I have an idea. I'm going to go cut up some code. I get to work. I forget about it and move on with life. I come back six months or a year or two years or three years later, you dust it off and you go use it again in a new environment. And maybe your GP is different. Maybe you're running on a server instead of a laptop, maybe whatever. Right. And so the problem now is you say, OK, well, I mean, again, not everybody cares about performance, but if you do, you say, OK, well, I want to take advantage of all these new features. I don't want to break the old thing, though. Right. And so the typical way of handling this kind of stuff before is, you know, if you're talking about people's templates, or you're talking about C with macros, you end up with if deaths, you get like all these weird things get layered in, make the code super complicated. And then how do you test it becomes this crazy complexity, multidimensional space you have to worry about? And, you know, that just doesn't scale very well." }, { "from": "human", "value": "Actually, let me just jump around before I go to some specific features like the increase in performance here that we're talking about can be just insane. You write that Moja can provide a 35000 X speed up over Python." }, { "from": "gpt", "value": "How does it do that? How does it do that? Yeah, so I can even do more, but we'll get to that. So so first of all, when we say that we're talking about what's called C Python, it's the default Python that everybody uses when you type Python three. That's like typically the one you use. Right. So C Python is an interpreter. And so interpreters, they have an extra layer of bike codes and things like this that they have to go read, parse, interpret, and it makes them kind of slow from that perspective. And so one of the first things we do is we move to a compiler. And so I'm just moving to a compiler, getting the interpreter out of the loop is two to five to 10 X speed up depending on the code. So just out of the gate, like she's in more modern techniques. Now, if you do that, one of the things you can do is you can start to look at how CPython started to lay out data. And so one of the things that CPython did, and this isn't part of the Python spec necessarily, but this is just sets of decisions, is that if you take an integer, for example, it'll put it in an object because in Python, everything's an object. And so they do the very logical thing of keeping the memory representation of all objects the same. So all objects have a header, they have payload data, and what this means is that every time you pass around an object, you're passing around a pointer to the data. Well, this has overhead. It turns out that modern computers don't like chasing pointers very much and things like this. It means you have to allocate the data. It means you have to reference count it, which is another way that Python uses to keep track of memory. And so this has a lot of overhead. And so if you say, okay, pointer, let's try to get that out of the heap, out of a box, of the new direction, and into the registers." } ]

[ { "from": "human", "value": "That's another 10x. So it adds up, if you reference counting every single thing you create, that adds up." }, { "from": "gpt", "value": "That adds up. Yeah, and if you look at people complain about the Python GIL, this is one of the things that hurts the parallelism, that's because of the reference counting, right, so the GIL and reference counting are very tightly intertwined in Python, it's not the only thing, but it's very tightly intertwined. And so then you lean into this and you say, OK, cool, well, modern computers, they can do more than one operation at a time. So they have vectors. What is a vector? Well, a vector allows you to take one, instead of taking one piece of data, doing an add or a multiply and then picking up the next one. You can now do 4, or 8, or 16, or 32 at a time. Well, Python doesn't expose that because of reasons. And so now you can say, OK, well, you can adopt that. And now you have threads. Now you have additional things, like you can control memory hierarchy. And so Moda allows you to do, is it allows you to start taking advantage of all these powerful things that have been built into the hardware over time and it gives, the library gives very nice features so you can say just parallelize this, do this in parallel, right? So it's very powerful weapons against slowness which is why people have been I think having fun like just taking code and making it go fast because it's just kind of an adrenaline" }, { "from": "human", "value": "rush to see like how fast you can get things. Before I talk about some of the interesting stuff with parallelization and all that, let's first talk about like the basics. We talked to indentation, right? So this thing looks like Python. It's sexy and beautiful like Python as I mentioned. Is it a typed language? So what's the role of types? Yeah, good question." }, { "from": "gpt", "value": "So yeah, good question. So Python has types, it has strings, it has integers, it has dictionaries and like all that stuff but they all live at runtime, right? And so because all those types live at runtime in Python you never, you don't have to spell them. Python also has like this whole typing thing going on now and a lot of people use it. I'm not talking about that. That's kind of a different thing. We can go back to that if you want. But typically the, you know, you just say I take, I have a def and my def takes two parameters and I'm gonna call them a and b and I don't have to write a type, okay? So that is great but what that does is that forces what's called a consistent representation. So these things have to be a pointer to an object with the object header and they all have to look the same. And then when you dispatch a method, when you go through all the same different paths no matter what the receiver, whatever that type is. So what Mojo does is it allows you to have more than one kind of type. And so what it does is allows you to say okay, cool. I have an object and objects behave like Python does, and so it's fully dynamic and that's all great. And for many things, classes like that's all very powerful and very important. But if you want to say hey it's an integer and it's 32 bits, or 64 bits, or whatever it is. Or it's a floating point value. It's 64 bits. Well then the compiler can take that and it can use that to do way better optimization. It turns out, again, getting rid of the indirection, that's huge, means you can get better code completion because compiler knows what the type is and so it knows what operations work on it. That's actually pretty huge. What Mojo does is it allows you to progressively adopt types into your program. You can start, again, it's compatible with Python. Then you can add however many types you want wherever you want them. And if you don't wanna deal with it you don't have to deal with it, right? And so one of our opinions on this is that it's not that types are the right thing" }, { "from": "human", "value": "or the wrong thing, it's that they're a useful thing. Which was kind of optional, it's not strict typing and you don't have to specify type. Exactly. Okay, so starting from the thing that Python is kind of reaching towards right now with trying to inject types into it," }, { "from": "gpt", "value": "what it's doing. Yeah, with a very different approach but yes." }, { "from": "human", "value": "Yeah, so. What's the different approach, I'm actually one of the people that have not been using types very much in Python. So I haven't. Why did you say? It just, well, because I know the importance. It's like adults use strict typing. And so I refuse to grow up in that sense. It's a kind of rebellion. But I just know that." }, { "from": "gpt", "value": "That's okay. Why did you say?" }, { "from": "human", "value": "But it probably reduces the amount of errors, even just for forget about performance improvements." }, { "from": "gpt", "value": "It probably reduces errors when you do strict typing. Yeah. So I mean, I think it's interesting if you look at that, right? The reason I'm giving you a hard time, is that there's this cultural norm, this pressure, there has to be a right way to do things. Growing up is only do it one way. And if you don't do that, you should feel bad. Some people feel like Python's a guilty pleasure or something. And that's like, when it gets serious, I need to go rewrite it. I mean, cool, I understand history and I understand kind of where this comes from, but I don't think it has to be a guilty pleasure. So if you look at that, you say, why do you have to rewrite it? Well, we have to rewrite to deploy. Well, why do you want to deploy? Well, you care about performance, or you care about perfectability or you want, a tiny thing on the server that has no dependencies or you know, you have objectives that you're trying to attain. So what if Python can achieve those objectives? So if you want types, well, maybe you want types because you want to make sure you're passing the right thing. Sure, you can add a type. If you don't care, you're prototyping some stuff, you're hacking some things out, you're like pulling some ram code off the internet, it should just work. And you shouldn't be pressured, you shouldn't feel bad about doing the right thing or the thing that feels good. Now, if you're in a team, you're working at some massive internet company and you have 400 million lines of Python code, well, they may have a house rule that you use types because it makes it easier for different humans to talk to each other and understand what's going on and bugs at scale. And so there are lots of good reasons why you might want to use types, but that doesn't mean that everybody should use them all the time. So what Mojo does is it says, cool, well, allow people to use types. And if you use types, you get nice things out of it, right? You get better performance and things like this, right? But Mojo is a full, compatible super set of Python, right? And so that means that has to work without types. It has to support all the dynamic things, has to support all the packages, it has to support list comprehensions and things like this, right? And so that that starting point, I think, is really important. And I think that, again, you can look at why I care so much about this, and there's many different aspects of that, one of which is the world went through a very challenging migration from Python 2 to Python 3, right? And this migration took many years, and it was very painful for many teams, right? And there's a lot of things that went on in that. I'm not an expert in all the details, and I honestly don't want to be. I don't want the world to have to go through that, right? And people can ignore Mojo, and if it's not their thing, that's cool. But if they want to use Mojo, I don't want them to have to rewrite all their code." } ]

[ { "from": "human", "value": "Yeah, I mean, okay, the superset part is just, I mean, there's so much brilliant stuff here. That definitely is incredible. We'll talk about that. First of all, how's the typing implemented differently in Python versus Mojo? So this heterogeneous flexibility, you said, is definitely implemented." }, { "from": "gpt", "value": "First of all, how is it definitely implemented? Yeah. So I'm not a full expert in the whole backstory on types in Python. So I'll give you that. I can give you my understanding. My understanding is basically, like many dynamic languages, the ecosystem went through a phase where people went from writing scripts to writing large scale, huge code bases in Python, and at scale, it kind of helps to have types. People want to be able to reason about interfaces. What do you expect to string or an int? These basic things, right? And so what the Python community started doing is it started saying, okay, let's have tools on the side, checker tools. Right? To go and enforce some variants, check for bugs, try to identify things. These are called static analysis tools generally, and so these tools run over your code and try to look for bugs. What ended up happening is there's so many of these things, so many different weird patterns and different approaches on specifying the types and different things going on that the Python community realize and recognize, hey, hey, hey, there's a thing here. And so what they started to do is they started to standardize the syntax for adding types to Python. Now, one of the challenges that they had is if they're coming from kind of this fragmented world where there's lots of different tools, they have different trade-offs and interpretations and the types mean different things. And so, if you look at types in Python, according to the Python spec, the types are ignored, right? So, according to the Python spec, you can write pretty much anything in a type position, okay? And technically, you can write any expression, okay. Now, that's beautiful because you can extend it. You can do cool things. build your own tools, you can build your own house, linter or something like that, right? But it's also a problem because any existing Python program may be using different tools and they have different interpretations. And so if you adopt somebody's package into your ecosystem, try around the tool you prefer, it may throw out tons of weird errors and warnings and problems just because it's incompatible with how these things work. Also because they're added late and they're not checked by the Python interpreter, it's always kind of more of a hint than it is a requirement. Also the CPython implementation can't use them for performance and so it's really-" }, { "from": "human", "value": "That's a big one, right? So you can't utilize for the compilation," }, { "from": "gpt", "value": "for the just in time compilation. Okay. Exactly. And this all comes back to the design principle of it's, they're kind of hints, they're kind of the definitions a little bit murky. It's unclear exactly the interpretation in a bunch of cases. And so because of that, you can't actually, even if you want to, it's really difficult to use them to say like, it is going to be an int and if it's not, it's a problem, right? A lot of code would break if you did that. So in Mojo, right. So you can still use those kinds of type annotations, it's fine, but in Mojo if you declare a type and you use it, then it means it is going to be that type and the compiler helps you check that enforce it and it's safe." }, { "from": "human", "value": "And it's not, it's not a best effort hint kind of a thing. So if you try to shove a string type thing into an integer- You get an error. And you get another." }, { "from": "gpt", "value": "From the compiler at compile time. You get an error." }, { "from": "human", "value": "And you get another. Nice." }, { "from": "gpt", "value": "Okay, what kind of basic types are there? Yeah, so Mojo is pretty hardcore in terms of what it tries to do in the language, which is the philosophy there is that we, again, if you look at Python, Python's a beautiful language because it's so extensible. And so all of the different things in Python, like for loops and plus, and all these things can be accessed through these underbar and bar methods. Okay, so you have to say, okay, if I make something that is super fast, I can go all the way down to the metal. Why do I need to have integers built into the language? And so what Mojo does is it says, okay, well, we can have this notion of structs. So you have classes in Python, now you can have structs. Classes are dynamic, structs are static. Cool, we can get high performance, we can write C++ kind of code with structs if you want. These things mix and work beautifully together. But what that means is that you can go and implement strings and ints and floats and arrays and all that kind of stuff in the language, right? And so that's really cool because, you know, to me as a idealizing compiler language type of person, what I wanna do is I wanna get magic out of the compiler and put it in the libraries. Because if somebody can, you know, if we can build an integer that's beautiful and it has an amazing API and it does all the things you'd expect an integer to do, but you don't like it, maybe you want a big integer, maybe you wanna like sideways integer, I don't know, like what all the space of integers are, then you can do that and it's not a second-class citizen. And so if you look at certain other languages like C++, one I also love and use a lot, int is hard-coded in the language, but complex is not. And so it's kind of weird that, you know, you have this STD complex class, but you have int. And complex tries to look like a natural numeric type and things like this, but integers in floating point have these like special promotion rules and other things like that, they're magic and they're hacked into the compiler. And because of that, you can't actually make something" } ]

[ { "from": "human", "value": "that works like the built in types. Is there something provided as a standard, because you know, because it's AI first, you know, numerical types is so important here, so is there something come like a nice standard implementation of integer inflow?" }, { "from": "gpt", "value": "Yeah, yeah. So we're still building all that stuff out, so we provide entries and floats and all that kind of stuff. We also provide buffers and tensors and things like that that you'd expect in an ML context. Honestly, we need to keep designing and redesigning and working with the community to build that out and make that better. That's not our strength right now. Give us six months or a year, and I think it'll be way better. But the power of putting in the library means that we can have teams of experts that aren't" }, { "from": "human", "value": "compiler engineers that can help us design and refine and drive us forward. So one of the exciting things we should mention here is that this is new and fresh. This cake is unbaked. It's almost baked. You can tell it's delicious, but it's not fully ready to be consumed." }, { "from": "gpt", "value": "Yep. That's very fair. It is very useful, but it's very useful if you're a super low-level programmer right now. And what we're doing is we're working our way up the stack. And so the way I would look at Mojo today in May in 2023 is that it's like a 0.1. So I think that a year from now, it's going to be way more interesting to a variety of people. But what we're doing is we decided to release it early so that people can get access to and play with it, and we can build it with the community. We have a big roadmap, fully published, being transparent about this. And a lot of people are involved in this stuff. And so what we're doing is we're really optimizing for building this thing the right way and building it the right way. It's interesting working with the community, because everybody wants it yesterday." }, { "from": "human", "value": "And so sometimes it's kind of, you know, there's some dynamics there, but I think it's the right thing. So there's a discord also. So the dynamics is pretty interesting. Sometimes the community probably can be very chaotic and introduce a lot of stress. Guido famously quit over the stress of the wall of his operator. I mean, it broke, maybe that would be, yeah, exactly. And so like, it could be very stressful to develop, but can you just tangent upon a tangent? Is it stressful to work through the design of various features here, given that the community is so richly involved?" }, { "from": "gpt", "value": "Yeah. Yeah. I mean, it's, you know, it broke. It probably broke the camel's back. Exactly. So I've been doing open development and community stuff for decades now. Somehow this has happened to me. So I've learned some tricks, but the thing that always gets me is I want to make people happy. Right? And so this is maybe not all people all happy all the time, but generally I want people to be happy. Right? And so the challenge is that, again, we're tapping into some long, some deep seated long tensions and pressures, both in the Python world, but also in the AI world and the hardware world and things like this. And so people just want us to move faster, right? And so again, our decision was let's release this early. Let's get people used to it or access to it and play with it and like, let's build in the open, which we could have had the language monk sitting in the cloister up on the hilltop, like beavering away trying to build something. But in my experience, you get something that's way better if you work with the community. And so yes, it can be frustrating and can be challenging for lots of people involved. And, you know, if you, I mean, if you mentioned our Discord, we have over 10,000 people on the Discord, 11,000 people or something. Keep in mind we released Mojo like two weeks ago. So it's very active. So it's very cool. But what that means is that, you know, 10, 11,000 people all will want something different. And so what we've done is we've tried to say, okay, cool, here's our roadmap. The roadmap isn't completely arbitrary. It's based on here's the logical order in which to build these features or add these capabilities and things like that. And what we've done is we've spun really fast on bug fixes. And so we actually have very few bugs, which is cool. I mean, actually for a project in the state, but then what we're doing is we're dropping in features very deliberately. I mean, this is cool." }, { "from": "human", "value": "I mean, this is fun to watch because you got the two gigantic communities of like hardware like systems engineers, and then you have the machine learning Python people that are like higher level. Yeah." }, { "from": "gpt", "value": "And it's just too like for like army like they've been at war. Yeah. Yeah. Sure. Right. And so here's a test. It's like a Tolkien novel or something. Okay. So here's the test. And again, like it's super funny for something that's only been out for two weeks, right? People are so impatient, right? But okay, cool. Let's fast forward a year. Like in a year's time, Mojo will be actually quite amazing and solve tons of problems and be very good. People still have these problems, right? And so you look at this and you say, and the way I look at this at least is to say, okay," }, { "from": "human", "value": "well, we're solving big, longstanding problems, right? And so here's a test. It's like a Tolkien novel or something." }, { "from": "gpt", "value": "Okay. To me, I, again, working on many different problems, I want to make sure we do it right. Right? There's like a responsibility feel because if you mess it up, right, there's very few opportunities to do projects like this and have them really have impact on the world. If we do it right, then maybe we can take those feuding armies and actually heal some of those wounds." } ]

[ { "from": "human", "value": "Yeah. Right? Yeah. I mean, this is like a speech by George Washington or Abraham Lincoln or something." }, { "from": "gpt", "value": "And you look at this and it's like, okay, well, how different are we? Yeah. We all want beautiful things. We all want something that's nice. We all want to be able to work together. We all want our stuff to be used, right? And so if we can help heal that, now I'm not optimistic that all people will use Mojo and they'll stop using C++. Like, that's not my goal. Right? But if we can heal some of that, I think that'd be pretty cool." }, { "from": "human", "value": "Yeah. Yeah. And we start by putting the people who like braces into the gulag. No." }, { "from": "gpt", "value": "So, so there are, so there are proposals for adding braces to Mojo and we just tell them no. Okay." }, { "from": "human", "value": "Delightly. Yeah. Anyway, so there's a lot of amazing features on the roadmap and those ready to implement it. It'd be awesome. I can just ask you a few things. Yeah. Go for it. So, the, the other performance improvement comes from immutability. So what's the, what's this var and this let thing that we got going on?" }, { "from": "gpt", "value": "What's immutability? Yeah. So, one of the things that is useful and it's not always required, but it's useful is knowing whether something can change out from underneath you. Right? And so in Python you have a pointer to an array, right? And so you pass that pointer to an array around to things. If you pass into a function, they may take that and scroll away in some other data structure. And so you get your array back and you go to use it and now somebody else is like putting stuff in your array. Yeah. How do you reason about that? It can be very complicated and at least lots of bugs, right? And so one of the things that, you know, again, this is not something Mojo forces on you, but something that Mojo enables is a thing called value semantics. And what value semantics do is they take collections like arrays, like dictionaries, also tensors and strings and things like this that are much higher level and make them behave like proper values. And so it makes it look like if you pass these things around, you get a logical copy of all the data. And so if I pass you an array, it's your array. You can go do what you want to it. You're not going to hurt my array. Now that is an interesting and very powerful design principle, defines away a ton of bugs. You have to be careful to implement it in an efficient way." }, { "from": "human", "value": "Yeah." }, { "from": "gpt", "value": "Is there a performance hit that's significant? Generally not. If you implement it the right way, but it requires a lot of very low level getting the language right bits." }, { "from": "human", "value": "I assume there'd be a huge performance hit because it's a really, the benefit is really" }, { "from": "gpt", "value": "nice because you don't get into the complex. Absolutely. Well, the trick is you can't do copies. So you have to provide the behavior of copying without doing the copy. Yeah." }, { "from": "human", "value": "How do you do that?" }, { "from": "gpt", "value": "How do you do that? It's not magic. It's just, it's actually pretty cool. Well, so first, before we talk about how that works, let's talk about how it works in Python. So in Python, you define a person class or maybe a person class is a bad idea. You define a database class, right? The database class has an array of records, something like that. And so the problem is, is that if you pass in a record or a class instance into database, it'll take a hold of that object and then it assumes it has it. And if you're passing an object in, you have to know that that database is going to take it. And therefore you shouldn't change it after you put in the database. This is the problem. You just kind of have to know that. You just have to kind of know that, right? And so you roll out version 1 of the database, you just kind of have to know that. Of course, Lex uses its own database, right? Yeah. Right? Because you built it. You understand how this works, right? Somebody else joins the team. They don't know this. Yes. Right? And so now they suddenly get bugs. You're having to maintain the database. You shake your fist. Mm-hmm. You argue the tenth time this happens. You're like, okay, we have to do something different, right? And so what you do is you go change your Python code and you change your database class to copy the record every time you add it. And so what ends up happening is you say, okay, I will do what's called a defensive copy inside the database. And then that way, if somebody passes something in, I will have my own copy of it. And they can go do whatever, and they're not going to break my thing. Okay. This is usually the two design patterns. If you look in PyTorch, for example, this is cloning a tensor. There's a specific thing, and you have to know where to call it. And if you don't call in the right place, you get these bugs, and this is state of the art, right? Right? Yeah. So a different approach. So it's used in many languages. So I work with it in Swift. As you say, okay, well, let's provide value semantics. And so we want to provide the view that you get a logically independent copy. But we want to do that lazily. And so what we do is we say, okay, if you pass something into a function, it doesn't actually make a copy. What it actually does is it just increments the reference to it. And if you pass it around, you stick in your database, they can go into the database. You own it. And then you come back out of the stack, nobody has copied anything. You come back out of the stack and then the caller lets go of it. Well, then you've just handed it off to the database. you've transferred it, and there's no copies made. Now on the other hand, if your coworker goes and hands you a record and you pass it in, you stick it in the database, and then you go to town and you start modifying it, what happens is you get a copy lazily on demand. And so what this does is gives you copies only when you need them, so it defines away" } ]

[ { "from": "human", "value": "the bugs, but also generally reduces the number of copies in practice. But the implementation details are tricky here? Yeah. So this is, yes. Sometimes it was reference counting, but to make it performant across a number of different kinds of objects." }, { "from": "gpt", "value": "Yeah. Well, so you need a couple of things, and so there's many, so this concept has existed in many different worlds, and again, it's not novel research at all, right? The magic is getting the design right so that you can do this in a reasonable way, right? And so there's a number of components that go into this. One is when you're passing around, so we're talking about Python and reference counting and the expense of doing that, when you're passing values around, you don't want to do extra reference counting for no good reason, and so you have to make sure that you're efficient and you transfer ownership instead of duplicating references and things like that, which is a very low-level problem. You also have to adopt this, and you have to build these data structures. And so if you say, you know, Mojo has to be compatible with Python, so of course the default list is a reference semantic list that works the way you'd expect in Python, but then you have to design a value semantic list, and so you just have to implement that and then you implement the logic within. And so the role of the language here is to provide all the low-level hooks that allow the author of the type to be able to get and express this behavior without forcing it into all cases or hardcoding this into the language itself. But there's ownership so you're constantly tracking who owns the thing, who owns the thing? Yes, and so there's a whole system called ownership, and so this is related to work done in the Rust community, also the Swift community has done a bunch of work and there's a bunch of different other languages that have all kind of C++ actually has copy constructors and destructors and things like that. And so, and, I mean, C++ has everything so it has move constructors, and has like this whole world of things and so this is, this is a body of work that's kind of been developing for many years now. And so Mojo takes some of the best ideas out of all these systems and remixes it in a nice of ways, so that you get the power of something like the Rust programming language. But you don't have to deal with it when you don't want to, which is a major thing in terms" }, { "from": "human", "value": "of teaching and learning and being able to use and scale these systems. How does that play with argument – conventions? What are they? Why are they important? How does the Bayley semantics, how does the transfer ownership, work with the arguments" }, { "from": "gpt", "value": "when they're pastened? Yeah, so if you go deep into a systems programming land – this isn't, again, this is not something for everybody, but if you go deep into systems-programming land, what you encounter is you encounter these types that get weird. So if you're used to Python, you think about everything, I can just copy it around, I can go change it and mutate it, and do these things, and it's all cool. If you get into systems-programming land, you get into these things like I have an atomic number, or I have a mutex, or I have a uniquely-owned database handle, things like this, right? these types you can't necessarily copy. Sometimes you can't necessarily even move them to a different address. And so what Mojo allows you to do is it allows you to express, hey, I don't want to get a copy of this thing. I want to actually just get a reference to it. And by doing that, well, you can say, you can say, okay, if I'm defining something weird, like a atomic number or something, it's like it has to be, so an atomic number is a, an area in memory that multiple threads can access at a time without synchronous, without locks. And so and so like the definition of atomic numbers, multiple, different things have to be poking up that therefore they have to agree on where it is. And he can't just like move it out from underneath one because it kind of breaks what it means. And so that's an example of a type that you can't even you can't copy. You can't move it like once you created it has to be where it was. Now if you look at many other examples like a database handle, right? So, okay, well, what happens? How do you copy a database handle? Do you copy the whole database? That's not something you necessarily want to do. There's a lot of types like that where you want to be able to say that they are uniquely owned. So, there's always one of this thing and or if I create a thing, I don't copy it. And so, what Mojo allows you to do is it allows you to say, hey, I want to pass around a reference to this thing without copying it. And so, it has borrowed conventions. So, you can say you can use it, but you don't get to change it. You can pass it by mutable reference. So, if you do that, then you can get a reference to it, but you can change it. And so, it manages all that kind of stuff." }, { "from": "human", "value": "So, this is just a really nice implementation of make a C++ has, you know, the- Yes, reference has- The categories. Yes. Smart, smart- different kinds of implementations and smart pointers that you can explicitly define this a lot, but you're saying that's more like" }, { "from": "gpt", "value": "the weird case versus the common case. The different kinds of pointers. Yeah, it has pointers. Smarts. Well, it depends on where, I mean, I don't think I'm a normal person, so I mean, I'm not one to call other people weird, but if you talk to a typical Python programmer, you're typically not thinking about this, right? This is a lower level of abstraction. Now, if you talk to a C++ programmer, certainly if you talk to a Rust programmer, again, they're not weird, they're delightful. Like these are all good people, right? Those folks will think about all the time, right? And so I look at this as there's a spectrum between very deep low level systems. I'm gonna go poke the bits and care about how they're laid out in memory, all the way up to application and scripting and other things like this. And so it's not that anybody's right or wrong," } ]

[ { "from": "human", "value": "it's about how do we build one system that scales." }, { "from": "gpt", "value": "I don't, so." }, { "from": "human", "value": "Yes, yeah. By the way, the idea of an atomic number has been something that always brought me deep happiness because the flip side of that, the idea that threads can just modify stuff asynchronously, just the whole idea of concurrent programming" }, { "from": "gpt", "value": "is a source of infinite stress for me. Well, so this is where you jump into, again, you zoom out and get out of programming languages or compilers and you just look at what the industry has done. My mind is constantly blown by this, right? And you look at Moore's law. Moore's law has this idea that computers, for a long time, single thread performance just got faster and faster and faster and faster for free. But then physics and other things intervened and power consumption, other things started to matter. And so what ended up happening is we went from single core computers to multi-core, then we went to accelerators. This trend towards specialization of hardware is only gonna continue. And so for years, us programming language nerds and compiler people have been saying, okay, well, how do we tackle multi-core, right? For a while, it was like, multi-core is the future. We have to get on top of this thing. And then it was multi-core is the default. What are we doing with this thing? And then it's like, there's chips with hundreds of cores in them, what happened, right? And so I'm super inspired by the fact that in the face of this, those machine learning people invented this idea of a tensor, right, and what is a tensor? A tensor is an, like an arithmetic and algebraic concept. It's like an abstraction around a gigantic, parallelizable data set, right? And because of that, and because of things like TensorFlow and PyTorch, we're able to say, okay, well, express the math of the system. This enables you to do automatic differentiations, enables you to do like all these cool things. And it's an abstract representation. Because you have that abstract representation, you can now map it onto these parallel machines without having to control, okay, put that right here, put that right there, put that right there. And this has enabled an explosion in terms of AI, compute, accelerators, like all the stuff. And so that's super, super exciting." }, { "from": "human", "value": "What happened, right? What about the deployment, the execution across multiple machines? So you write that the modular compute platform dynamically partitions models with billions of parameters and distributes their execution across multiple machines, enabling unparalleled efficiency. By the way, the use of unparalleled in that sentence, anyway, enabling unparalleled efficiency scale and the reliability for the largest workloads. So how do you do this abstraction" }, { "from": "gpt", "value": "of distributed deployment of large? Yeah, so one of the really interesting tensions. So there's a whole bunch of stuff that goes into that. I'll pick a random walk through it. If you go back and replay the history of machine learning, right, I mean, the brief, the brief, the most recent history of machine learning, because this is, as you know, very deep. I knew Lex when he had an AI podcast. Yes. Right, yep. So if you look at just TensorFlow and PyTorch, which is pretty recent history in the big picture, right? But TensorFlow is all about graphs. PyTorch, I think pretty unarguably ended up winning and why did it win? Mostly because of usability, right? And the usability of PyTorch is, I think, huge. And I think, again, that's a huge testament to the power of taking abstract, theoretical, technical concepts and bringing it to the masses, right? Now, the challenge with what the TensorFlow versus the PyTorch design points was that TensorFlow is kind of difficult to use for researchers, but it was actually pretty good for deployment. PyTorch is really good for researchers. It's kind of not super great for deployment, right? And so I think that we, as an industry, have been struggling. And if you look at what deploying a machine learning model today means is that you'll have researchers who are, I mean, wicked smart, of course, but they're wicked smart at model architecture and data and calculus. And they call it, they're wicked smart in various domains. They don't want to know anything about hardware or deployment or C++ or things like this, right? And so what's happened is you get people who train the model, they throw it over the fence and then you have people that tried to deploy the model. Well every time you have a team A does X, they throw it over the fence, and team B does Y, you have a problem. Because, of course, it never works the first time. And so, you throw it over the fence, they figure out, okay, it's too slow, won't fit, doesn't use the right operator, the tool crashes, whatever the problem is, then they have to throw it back over the fence. And every time you throw a thing over a fence, It takes three weeks of project managers and meetings and things like this. And so, what we've seen today is that getting models in production can take weeks or months. Like, it's not atypical, right? I talk to lots of people and you talk about like, VPS software at some internet company trying to deploy a model, and they're like, why do I need a team of 45 people? It's so easy to train a model. Why can't I deploy it, right? And if you dig into this, every layer is problematic. So if you look at the language piece, I mean, this is the tip of the ice bird, it's a very exciting tip of the iceberg for folks, but you've got Python on one side and C++ on the other side. Python doesn't really deploy. I mean, it can theoretically, technically in some cases, but often a lot of production teams will want to get things out of Python because they get their performance and control and whatever else. So Mojo can help with that. If you look at serving, so you talk about gigantic models. Well, a gigantic model won't fit on one machine, right? And so now you have this model, it's written in Python, it has to be rewritten in C++. Now it also has to be carved up so that half of it runs on one machine, half of it runs on another machine, or maybe it runs on 10 machines. Well, so now suddenly the complexity is exploding, right? And the reason for this is that if you look into TensorFlow, PyTorch, these systems, they weren't really designed for this world, right? They were designed for, you know, back in the day when we were starting and doing things where it was a different, much simpler world, like you want to run ResNet-50 or some ancient model architecture like this, and it was just a completely different world. Train on one GPU, doing something on one GPU. Yeah, AlexNet, right? In the major breakthrough. And the world has changed, right? And so now the challenge is that TensorFlow, PyTorch, these systems, they weren't actually designed for LLMs, like that was not a thing. And so where TensorFlow actually has amazing power in terms of scale and deployment, and things like that, and I think Google is, I mean, maybe not unmatched, but they're like incredible in terms of their capabilities and gigantic scale. Many researchers using PyTorch, right? And so PyTorch doesn't have those same capabilities. And so Modular can do is it can help with that. Now, if you take a step back and say, like, what is Modular doing, right? So Modular has like a AlexNet, a bitter enemy that we're fighting against in the industry. And it's one of these things where everybody knows it, but nobody is usually willing to talk about it. The bitter enemy. The bitter thing that we have to destroy, that we're all struggling with, and it's like fish can't see water, is complexity. Yes, it's complexity. That was very close off. And so if you look at it, yes, it is on the hardware side. Yes. All these accelerators, all these software stacks that go with the accelerator, all these, there's just massive complexity over there. You look at what's happening on the modeling side, massive amount of complexity. Like things are changing all the time. People are inventing... Turns out the research is not done, right? And so people want to be able to move fast. Transformers are amazing, but there's a ton of diversity even within Transformers. And what's the next Transformer, right? And you look into serving. Also huge amounts of complexity. It turns out that all the cloud providers have all their very weird, but very cool hardware. But very cool hardware for networking and all this kind of stuff and it's all very complicated people aren't using that you look at Classical serving right there There's this whole world of people who know how to write high-performance servers with zero copy networking and like all all this fancy Asynchronous IO and like all these fancy things in the in the serving community Very little that has pervaded into the machine learning world Right, and why is that? Well, it's because again these systems have been built up over many years They they haven't been rethought there hasn't been a first principles approach to this. And so modular is doing as we're saying, okay" } ]

[ { "from": "human", "value": "Complexly" }, { "from": "gpt", "value": "We've built many of these things like so I've worked on TensorFlow and TPUs and things like that other folks on our team like worked on PyTorch core We've worked on onyx one time we've worked on many of these other systems and so built The systems like the Apple accelerators and all that kind of stuff like our team is quite amazing. And so one of the things that Roughly every modular is grumpy about is that when you're working on one of these projects you have a first-order goal Get the hardware to work get the system to enable one more model get this product out the door enable the specific workload or make it Solve this problem for this this product team, right and nobody's been given a chance to actually do that step back And so we as an industry we didn't take two steps forward we took like 18 steps forward in terms of all this really cool technology across compilers and systems and run times and heterogeneous computing like all this kind of stuff and like all this technology has been You know, I wouldn't say Beautifully designed but it's been proven in different quadrants like, you know, you look at Google with TPUs Massive huge exa flops of compute strapped together and into machines that researchers are programming in Python in a notebook That's huge. It's amazing. That's incredible, right? It's incredible. And so you look at the technology that goes into that and The the algorithms are actually quite general and so Lots of other hardware out there and lots of other teams out there don't have the sophistication or that maybe the the years working on it Or the the budget or whatever that Google does, right? And so they should be getting access to same algorithms, but they just don't have that and it's what modular is doing So saying it's just amazing cool. This is not research anymore. Like we've we've built auto tuning in many systems We've built programming languages, right? And so I kept have you know implemented c++ have implemented Swift have implemented many of these things And so, you know this it's hard, but it's not research and you look at accelerators Well, we know there's a bunch of different weird kind of accelerators, but they actually cluster together Right and you look at GPUs. Well, there's a couple of major vendors of GPUs and they maybe don't always get along But their architectures are very similar you look at CPUs CPUs are still super important for the deployment side of things You see new new architectures coming out from all the cloud providers and things like this and they're all super important to the world All right But they don't have the 30 years of development that the entrenched people do right and so what modular can do is we're saying Okay, all this complexity. Like it's not it's not bad complexity. It's actually innovation Right. And so it's innovation that's happening and it's for good reasons But I have sympathy for the poor software people, right? I mean again, i'm a generally a software person, too I love hardware but software people want to build applications and products and solutions that scale over many years They don't want to build a solution for one generation of hardware with one vendors tools And because of this they need something that scales with them. They need, something that works on cloud and mobile, and so I like it because their product manager said, hey, I want it to be, to have lower latency, and it's better for personalization, or whatever they decide, right. Products evolve and so the challenge with the machine learning technology and the infrastructure that we have today in the industry is that it's all these point solutions. And because there are all these point solutions, it means that as your product evolves you have to, to like switch different technology stacks or switch to different vendor." }, { "from": "human", "value": "And what that does is that slows down progress. So basically, a lot of the things we've developed in those little silos for machine learning tasks, you want to make that the first-class citizen of a general purpose programming language that can then be compiled" }, { "from": "gpt", "value": "across all these kinds of hardware. Well, so it's not really about a programming language. I mean, the programming language is a component of the mission, right? And the mission is, or not literal, but our joking mission is to save the world from terrible AI software. Excellent, I love this. So, you know, if you look at this mission, you need a syntax. So that's, so yeah, she needed programming language, right? And like we wouldn't have to build the programming language if one existed, right? So if Python was already good enough, then cool. We would just used it, right? We're not just doing very large scale, expensive engineering projects for the sake of it. Like it's to solve a problem, right? It's also about accelerators. It's also about exotic numerics and B float 16 and matrix multiplications and convolutions and like this kind of stuff. Within the stack, there are things like kernel fusion. That's an esoteric, but really important thing that leads to much better performance and much more general research hackability together, right?" }, { "from": "human", "value": "And that's enabled by the ASICs. That's enabled by certain hardware. Well, it's like, where's the dance between, I mean, there's several questions here. Like, how do you add a piece of hardware to this stack? If I have this genius invention of a specialized accelerator, how do I add that to the modular framework? And also how does modular as a standard start to define the kinds of hardware" }, { "from": "gpt", "value": "that should be developed? Yeah, so let me take a step back and talk about status quo, okay? And so if you go back to TensorFlow One, PyTorch One, this kind of timeframe, and these have all evolved and gotten way more complicated. So let's go back to the glorious simple days, right? These things basically were CPUs and CUDA. And so what you do is you say, go do a dense layer and a dense layer has a matrix multiplication in it, right? And so when you say that, you say, go do this big operation of matrix multiplication. And if it's on a GPU, kick off a CUDA kernel. If it's on a CPU, go do like an Intel algorithm or something like that with the Intel MKO, okay? Now, that's really cool if you're either video or Intel, right, but then more hardware comes in, right? And on one axis, you have more hardware coming in. On the other hand, you have an explosion of innovation in AI. And so what happened with both TensorFlow and PyTorch is that the explosion of innovation in AI has led to, it's not just about matrix multiplication and convolution, these things have now like 2000 different operators. And on the other hand, you have, I don't know how many pieces of hardware there are out there, it's a lot, okay? It's not even hundreds, it's probably thousands, okay? And across all of Edge and across like all the different things. That are used at scale." } ]

[ { "from": "human", "value": "Yeah, exactly, I mean, AI's everywhere, yeah. Yeah, exactly, I mean, AI's everywhere, yeah." }, { "from": "gpt", "value": "It's not a handful of TPU alternatives. Correct. It's every phone often with many different chips inside of it from different vendors, right?" }, { "from": "human", "value": "Like it's, AI is everywhere, it's a thing, right? Why are they all making their own chips? Like why is everybody making their own thing? Well, so because-" }, { "from": "gpt", "value": "Is that a good thing, for sure? So Chris's philosophy on hardware, right? So my philosophy is that there isn't one right solution, right, and so I think that, again, we're at the end of Moore's law, specialization happens. If you're building, if you're training GPT-5, he wants some crazy supercomputer data center thingy. If you're making a smart camera that runs on batteries, you want something that looks very different. If you're building a phone, you want something that looks very different. If you have something like a laptop, you want something that looks, maybe similar but a different scale, right? And so AI ends up touching all of our lives, robotics, right, and like lots of different things. And so as you look into this, these have different power envelopes, There's different trade-offs in terms of the algorithms. There's new innovations in sparsity and other data formats and things like that. And so hardware innovation, I think is a really good thing, right? And what I'm interested in is unlocking that innovation. There's also like analog and quantum and like all the really weird stuff, right? And so if somebody can come up with a chip that uses analog computing and it's a hundred X more power efficient, think what that would mean in terms of the daily impact on the products we use. That'd be huge. Now, if you're building an analog computer, you may not be a compiler specialist. The really weird stuff, right? These are different skill sets, right? And so you can hire some compiler people if you're running a big company maybe. But it turns out these are really like exotic new generation of compilers. Like this is a different thing, right? And so if you take a step back out and come back to what is the status quo? Status quo is that if you're Intel or you're Nvidia, you continue to keep up with the industry and you chase and okay, there's 1900, now there's 2000, now there's 2100 and you have a huge team of people that are trying to keep up in tune and optimize. And even when one of the big guys comes out with a new generation of their chip, they have to go back and rewrite all these things, right? So really it's only powered by having hundreds of people that are all like frantically trying to keep up. And what that does is that keeps out the little guys and sometimes they're not so little guys, the big guys are also just not in those dominant positions. And so what has been happening, and so a lot of, you talk about the rise of new exotic crazy accelerators, is people have been trying to turn this from a, let's go write lots of special kernels problem into a compiler problem. And so we and I contributed to this as well. We as an industry went into it like, let's go make this compiler problem phase, let's call it. And much of the industry is still in this phase, by the way, so I wouldn't say this phase is over. And so the idea is to say, look, okay, what a compiler does is it provides a much more general, extensible, hackable interface for dealing with the general case, right? And so within machine learning algorithms, for example, people figured out that, hey, if I do a matrix multiplication and I do a ReLU, right? The classic activation function, it is way faster to do one pass over the data and then do the ReLU on the output where I'm writing out the data, because ReLU is just a maximum operation, right? Maximum zero. And so it's an amazing optimization, take matmul ReLU squished together in one operation, now I have matmul ReLU. Well, wait a second. If I do that now, I just went from having, two operators to three, but now I figure out, okay, well, there's a lot of activation functions. What about leaky ReLU? What about like a million things that are out there, right? And so as I start fusing these in, now I get permutations of all these algorithms, right? And so with the compiler people said, is they said, hey, cool. Well, I will go enumerate all the algorithms and I will enumerate all the pairs and I will actually generate a kernel for you. And I think that this has been very, very useful for the industry. This is one of the things that powers Google TPUs, pyTorch2s, like rolling out really cool compiler stuff with Triton and this other technology and things like this. And so the compiler people are kind of coming into their fore and saying like awesome, It's awesome! This is a compiler problem,REL compiler it.\" Right? Here's the problem, not everybody's a compiler person. I love compiler people, trust me. Right? But not everybody can or should be a compiler person. It turns out that there are people that know analog computers really well, or they know some GPU internal architecture thing really well, or they know some crazy sparse numeric interesting algorithm that is the cusp of research. But they're not compiler people. And so one of the challenges with this new wave of technology trying to turn everything into a compiler, because again, it has excluded a ton of people. And so you look at what does Mojo do? What does the modular stack do? This brings programmability back into this world. Like it enables, I wouldn't say normal people, but a different kind of delightful nerd that cares about numerics or cares about hardware or cares about things like this, to be able to express that in the stack and extend the stack without having to actually go" } ]

[ { "from": "human", "value": "hack the compiler itself. So extend the stack on the algorithm side," }, { "from": "gpt", "value": "and then on the hardware side. Yeah, so again, go back to the simplest example of int. And so what both Swift and Mojo and other things like this did is we said, OK, pull magic out of the compiler and put it in the standard library. And so what modular is doing with the engine that we're providing and this very deep technology stack, which goes into heterogeneous runtimes and a whole bunch of really cool things, this whole stack allows that stack to be extended and hacked and changed by researchers and by hardware innovators and by people who know things that we don't know because modular has some smart people," }, { "from": "human", "value": "but we don't have all the smart people, it turns out." }, { "from": "gpt", "value": "What are heterogeneous runtimes? Yeah, so what is heterogeneous, right? So heterogeneous just means many different kinds of things together. And so the simplest example you might come up with is a CPU and a GPU. And so it's a simple heterogeneous computer to say, I will run my data loading and preprocessing and other algorithms on the CPU. And then once I get it into the right shape, I shove it into the GPU. I do a lot of matrix multiplications and convolutions and things like this. And I get it back out and I do some reductions and summaries and I shove it across the network to another machine. And so you've got now what are effectively two computers, a CPU and a GPU talking to each other, working together in a heterogeneous system. But that was 10 years ago. You look at a modern cell phone. You've got CPUs. And they're not just CPUs. There's big dot little CPUs. And so there's multiple different kinds of CPUs that are, again, working together. They're multi-core. You've got GPUs. You've got neural network accelerators. You've got dedicated hardware blocks for media, so for video decode and JPEG decode and things like this. And so you've got this massively complicated system. And this isn't just cell phones. Every laptop these days is doing the same thing. And all these blocks can run at the same time and need to be choreographed, right? And so, again, one of the cool things about machine learning is it's moving things to like data flow graphs and higher level abstractions and tensors and these things that it doesn't specify. Here's how to do the algorithm. It gives the system a lot more flexibility in terms of how to translate or map or compile it onto the system that you have. And so what you need, you know, at the bottom-est part of the layer there is a way for all these devices to talk to each other. And so this is one thing that, you know, I'm very passionate about, I mean, you know, I'm a nerd. But all these machines and all these systems are effectively parallel computers running at the same time, sending messages to each other. And so they're all fully asynchronous. Well, this is actually a small version of the same problem you have in a data center, right? In a data center, you now have multiple different machines, sometimes very specialized, sometimes with GPUs or TPs in one node and sometimes with disks in another node. So you get a much larger scale heterogeneous computer. And so what ends up happening is you have this like multi-layer abstraction of hierarchical parallelism, hierarchical asynchronous communication. And making that, again, my enemy is complexity by getting that away from being different specialized systems at every different part of the stack and having more consistency and uniformity. I think we can help lift the world" }, { "from": "human", "value": "and make it much simpler and actually get used. But how do you leverage like the strengths of the different specialized systems? So looking inside the smartphone, like there's just, what, like, I don't know, five, six computers, essentially, inside the smartphone? How do you, without trying to minimize the explicit, making it explicit which computer is supposed to be for which operation?" }, { "from": "gpt", "value": "For which operation? Yeah, so there's a pretty well-known algorithm. And what you're doing is you're looking at two factors. You're looking at the factor of sending data from one thing to another, right? It takes time to get it from that side of the chip to that side of the chip and things like this. And then you're looking at what is the time it takes to do an operation on a particular block. So, take CPUs. CPUs are fully general. They can do anything. But then you have a neural net accelerator that's really good at matrix multiplications. And so, you say, okay, well, if my workload is all matrix multiplications, I start up, I send the data over the neural net thing, it goes and does matrix multiplications. when it's done, it sends me back the result. All is good. The simplest thing is just saying, do matrix operations over there. Then you realize you get a little bit more complicated, because you can do matrix multiplications on a GPU, you can do it on a neural net accelerator, you can do it on CPU, and they'll have different trade-offs and costs, and it's not just matrix multiplication. What you actually look at is you look at, I have generally a graph of compute, I want to do a partitioning, I want to look at the communication, the bisection bandwidth, and the overhead and the sending of all these different things, and build a model for this, and then decide, okay, it's an optimization problem, where do I want to place this compute?" }, { "from": "human", "value": "The old school theoretical computer science problem of scheduling. Then how does presumably" }, { "from": "gpt", "value": "it's possible to somehow magically include autotune into this? Absolutely. In my opinion, so this is an opinion, not everybody would agree with this, but in my opinion, the world benefits from simple and predictable systems at the bottom that you can control. But then, once you have a predictable execution layer, you can build lots of different policies on top of it. One policy can be that the human programmer says, do that here, do that here, do that here, do that here, and fully manually controls everything, and the system should just just do it. Then, you quickly get in the mode of, I don't want to have to tell it to do it. The next logical step that people typically take is they write some terrible heuristic. If it's a major location, do it over there. If it's floating point, do it on the GPU. If it's integer, do it on CPU, something like that. Then you then get into this mode of people care more and more and more, and you say, okay, well, let's actually make the heuristic better. Let's get into auto tuning. is actually do search of the space, to decide, well, what is actually better, right? Well, then you get into this problem where you realize this is not a small space, this is a many dimensional, us reasons dimensional space, that you cannot exhaustively search, so, do you know of any other algorithms that are good at searching very complicated spaces for-" } ]

[ { "from": "human", "value": "Don't tell me you're gonna turn this" }, { "from": "gpt", "value": "into a machine learning problem. So then you turn into a machine learning problem, And then you have a space of genetic algorithms and reinforcement learning and all these cool things." }, { "from": "human", "value": "Can you include that into the stack, into the modular stack? Yeah, yeah. Where does it sit? Where does it live?" }, { "from": "gpt", "value": "Is it a separate thing or is it part of the compilation? Yeah, yeah. So you start from simple and predictable models. And so you can have full control and you can have coarse-grained knobs like nudged systems, so you don't have to do this. But if you really care about getting the best, last ounce out of a problem then you can use additional tools and they're the cool things you don't want to do this every time you run a model you want to figure out the right answer and then cash it and once you do that you can get you can say okay cool I can get up and running very quickly I can get good execution out of my system I can decide if something's important and if it's important I can go throw a bunch of machines at it and do a big expensive search ever the space using whatever technique I feel like its pretty up to to the problem, and then when I get the right answer, cool, I can just start using it. And so you can get out of this trade-off between, okay, am I gonna spend forever doing a thing, or do I get up and running quickly, and as a quality result, these are actually not in contention with each other if the system's designed to scale." }, { "from": "human", "value": "You started and did a little bit of a whirlwind overview of how you get 35,000X speed up or more over Python. Jeremy Howard did a really great presentation about sort of the basic, like, look at the code, here's how you get the speed up. Like you said, that's something we could, probably developers can do for their own code to see how you can get this gigantic speed up. But can you maybe speak to the machine learning task in general, how do you make some of this code fast and specific, like, what would you say is the main bottleneck for machine learning tasks? So are we talking about that mall matrix multiplication, how do you make that fast?" }, { "from": "gpt", "value": "So, I mean, if you just look the Python problem, right, you can say, how do I make Python faster? There's been a lot of people that have been working on the, okay, how do I make Python 2X faster, 10X faster, or something like that, right? And there've been a ton of projects in that vein, right? Mojo started from the, what can the hardware do, like, what is the limit of physics? What is the speed of light, what is it? Like, how fast can this thing go? And then, how do I express that? Yeah, right. And so it wasn't well, it wasn't anchored relatively on make Python a little bit faster. It's saying cool I know what the hardware can do. Let's unlock that right now when you yeah" }, { "from": "human", "value": "Yeah, just say how how gutsy that is to be in the meeting and as opposed to trying to see how do we get the improvement?" }, { "from": "gpt", "value": "It's like what can the physics do? I Mean, maybe I'm a special kind of nerd, but you look at that. What is the limit of physics? How fast can these things go, right? When you start looking at that, typically it ends up being a memory problem. Mm-hmm. Right and so today Particularly with these specialized accelerators. The problem is that you can do a lot of math within them But yet you get bottleneck sending data back and forth to memory whether it be local memory or distant memory or disk or whatever It is and and that that bottleneck particularly is the training sizes get large as you start doing tons of inferences all over the place Like that becomes a huge bottleneck for people, right? So again, what happened is we went through a phase of many years where people took the special case and hand tuned it Tweaked it and tricked out and they knew exactly how the hardware worked and they knew the model and they made it they made it fast Didn't generalize and so you can make you know Resident 50 or some or Alex net or something inception v1 like you can you can do that, right because the models are small They fit in your head Right, but as the models get bigger more complicated as the machines get more complicated it stops working Right, and so this is where things like kernel fusion come in. So what is kernel fusion? This is this idea of saying let's avoid going to memory and let's do that by building a new hybrid Kernel and a numerical algorithm that actually keeps things in the accelerator instead of having to write all the way out to memory All right, what's happened with with these accelerators now is you get multiple levels of memory like in a GPU For example, you'll have global memory and local memory and like all these things If you zoom way into how hardware works the register file is actually a memory so the registers are like an L0 cache and so a lot of Taking advantage of the hardware ends up being full utilizing the full power in all of its capability and this has a number of problems, right one of which is again the complexity of disaster, right? There's too much hardware. Even if you just say let's look at The chips from one line of vendor like Apple or Intel or whatever it is Each version of the chip comes out with new features and they change things so that it takes more time or less time to do different Things and you can't rewrite all the software Whenever a new chip comes out, right? And so this is where you need a much more scalable approach and this is what mojo and what the Modular stack provides is it provides this infrastructure and the system for factoring all this complexity and then allowing people to express algorithms You talk about auto tuning for example Express algorithms in a more portable way so that when a new chip comes out you don't have to rewrite it all So to me like, you know, I kind of joke like what is a compiler? Well, there's many ways to explain that you convert thing a and thing B and you convert source code to machine code Like you can talk about many many Things that compilers do but to me it's about a bag of tricks It's about a system and a framework that you can hang Complexity it's a system that can then generalize and it can work on problems that are bigger than fit in one human's head Right. And so what that means what a good stack and what the modular stack provides is The ability to walk up to it with a new problem and it'll generally work quite well And that's something with a lot of machine learning infrastructure and tools and technologies don't have Typical state-of-the-art today as you walk up particularly if you're deploying if you walk up with a new model You try to push it through the converter and the converter crashes. That's crazy The state of ML tooling today it's not anything that a C programmer would ever accept, right? And it's always been this kind of flaky side of tooling that's never been integrated. Well, and it's been Never worked together and because it's not designed together. It's built by different teams It's built by different hardware vendors is built by different systems it's built by different internet companies They're trying to solve their their problems Right and so that means that we get this fragmented terrible mess of complexity" } ]

[ { "from": "human", "value": "So I mean the specifics of enemy Jeremy showed this. There's the vectorized function, which I guess is Built in to the into mojo Vectorized as he showed us built into the library into the library libraries in a library vectorized paralyzed Which vectorized is more low-level paralyzes higher level. There's the tiling thing Which is how he demonstrated the?" }, { "from": "gpt", "value": "Autotune I think so so think of think about this in like levels Hierarchical levels of abstraction right and so it at the very if you zoom all the way into a compute problem You have one floating point number right and so then you say okay I want to be I could do things one at a time in an interpreter It's pretty slow right so I can get to doing one one at a time in a compiler I can see then I can get to doing four or eight or sixteen at a time with vectors That's called vectorization Then you can say hey, I have a whole bunch of different You know what a multi-core computer is is this basically a bunch of computers? Right so they're all independent computers that they can talk to each other and they share memory And so now what parallelize does it says okay run multiple instances on different computers? And now they call work together our program right and so what you're doing is you're saying keep going out to the next level out And and as you do that. How do I take advantage of this so tiling is a memory optimization right It says okay Let's make sure that we're keeping the data close to the compute Part of the problem instead of sending it all back and forth through memory every every time. I would load a block And the size of the block size is all that's how you get to the auto tune to make right optimized right optimized Yeah, well so all of these the details matter so much to get good performance This is another funny thing about machine learning and high performance computing that is very different than See compilers we all grew up grew up with where you know If you get a new version of GCC or new version of clang or something like that, you know Maybe something will go 1% faster, right? And so compiler ensures will work really really really hard to get half a percent out of your C code something like that but when you're talking about an accelerator or an AI application or you're talking about these kinds of Algorithms and these are things people used to write in Fortran for example, right if you get it wrong It's not 5% or 1% that could be 2x or 10x Right if you think about it You really want to make use of the full memory you have the cache for example But if you use too much space it doesn't fit in the cache Now you're gonna be thrashing all the way back out to main memory and these can be 2x 10x Major performance differences and so this is where getting these magic numbers and these things right is really actually quite important" }, { "from": "human", "value": "So you mentioned that Mojo is a superset of Python. Can you run?" }, { "from": "gpt", "value": "Python code as if it's mojo code. Yes. Yes So and so and this has two sides of it. So mojo is not done yet. So I'll give you disclaimer mojo It's not done yet, but already we see people that take small pieces of Python code move it over They don't change it and you can get 12x speedups like somebody was just tweeting about that yesterday, which is pretty cool Right and again interpreters compilers, right? And so without changing a code without Also, this is not with this is not JIT compiling or do anything fancy. This is just Basic stuff move it straight over now mojo will continue to grow out and as it grows out It will have more and more and more features and our nor stars to be a full superset of Python And so you can bring over Basically arbitrary Python code and have it just work and it may not always be 12x faster But but it should be at least as fast and way faster in many cases. This is the goal, right? Now take time to do that and Python is a complicated language. There's not just the obvious things but there's also Non-obvious things that are complicated like we have to be able to talk to see Python Packages to talk to the C API and there's a bunch of there's a bunch of pieces you have not done yet. Cool" }, { "from": "human", "value": "You have to any just to make explicit the obvious I mean not be so obvious. I don't you think about it. So, you know to run Python code to run Python code, that means you have to run all the Python packages and libraries. Yeah. Yeah. So that means what? What's the relationship between Mojo and CPython, the interpreter that's presumably would be" }, { "from": "gpt", "value": "tasked with getting those packages to work and libraries to work? Yeah. Yep. So in the fullness of time, Mojo will solve for all the problems and you'll be able to move Python packages over and run them in Mojo. Without the CPython. I'll CPython someday. Yeah. Right. It's not today, but someday. And that'll be a beautiful day because then you'll get a whole bunch of advantages and you'll get massive speedups and things like this. So you can do that one at a time, right? You can move packages one at a time. Exactly. But we're not willing to wait for that. Python is too important. The ecosystem is too broad. We want to both be able to build Mojo out. We also want to do it the right way without time, like without intense time pressure. We're obviously moving fast, but, and so what we do is we say, okay, well, let's make it" } ]

[ { "from": "human", "value": "so you can import an arbitrary existing package." }, { "from": "gpt", "value": "You can do that one at a time, right? Carpetory, including like you write your own on your local disk or whatever it's not, it's not like an arbitrary package and import that using CPython, whatever it's not. CPython already runs all the packages. Right. And so what we do is we built an integration layer where we can actually use CPython. Again, I'm practical, and to actually just load and use all the existing packages as they are. The downside of that is you don't get the benefits of Mojo for those packages. They run as fast as they do in the traditional CPython way, but what that does is that gives you an incremental migration path. If you say, hey, cool, the Python ecosystem is vast. I want all of it to just work, but there's certain things that are really important. If I'm doing weather forecasting or something, well, I want to be able to load all the data. I want to be able to work with it, and then I have my own crazy algorithm inside of it. Normally, I'd write that in C++. If I can write in Mojo and have one system that scales, well, that's way easier to work" }, { "from": "human", "value": "with. Is it hard to do that, to have that layer that's running CPython, because is there some" }, { "from": "gpt", "value": "communication back and forth? Yes, it's complicated. This is what we do. We make it look easy, but it is complicated. What we do is we use the CPython existing interpreter, so it's running its own byte codes and that's how it provides full compatibility, and then it gives us CPython objects. We use those objects as is. That way, we're fully compatible with all the CPython objects and all the, it's not just the Python part, it's also the C packages, the C libraries underneath them, because they're often hybrid, so we can fully run and we're fully compatible with all that. The way we do that is that we have to play by their rules, so we keep objects in that representation when they're coming from that world. What's the representation that's being used in memory? We'd have to know a lot about how the CPython interpreter works. It has, for example, reference counting, but also different rules on how to pass pointers around and things like this. Super low-level fiddly, and it's not like Python. It's like how the interpreter works. And so that gets all exposed out, and then you have to define wrappers around the low-level C code. And so what this means is you have to know not only C, which is a different world from Python, obviously, not only Python, but the wrappers. But the interpreter, and the wrappers, and the implementation details, and the conventions, and it's just this really complicated mess. And when you do that, now suddenly you have a debugger that debugs Python. They can't step in the C code, so you have this two-world problem. And so by pulling this all into Mojo, what you get is you get one world. You get the ability to say, cool, you have untyped, very dynamic, beautiful, simple code. OK, I care about performance for whatever reason. There's lots of reasons you might care. And so then you add types. You can parallelize things. You can vectorize things. You can use these techniques, which are general techniques, to solve a problem. And then you can do that by staying in the system. And if you have that one Python package that's really important to you, you can move it to Mojo. You get massive performance benefits on that, and other advantages. If you like exact types, it's nice if they're enforced. Some people like that, rather than being hints. So there's other advantages, too." }, { "from": "human", "value": "And then you can do that incrementally as you go. So one different perspective on this would be, why Mojo instead of making C Python faster or redesigning C Python?" }, { "from": "gpt", "value": "Yeah, well, I mean, you could argue Mojo is redesigning C Python. But why not make C Python faster and better and other things like that? There's lots of people working on it. So actually, there's a team at Microsoft that is really improving. I think C Python 3.11 came out in October or something like that, and it was 15% faster, 20% faster across the board, which is pretty huge, given how mature Python is and things like this. And so that's awesome. I love it. Doesn't run on GPU. It doesn't do AI stuff. It doesn't do vectors. It doesn't do things. 20% is good. 35,000 times is better. So I'm a huge fan of that work, by the way, and it composes well with what we're doing. And so it's not like we're fighting or anything like that. It's actually just general goodness for the world. But it's just a different path. And again, we're not working forwards from making Python a little bit better." }, { "from": "human", "value": "We're working backwards from what is the limit of physics. What's the process of porting Python code to Mojo? Is there what's involved in that in the process?" }, { "from": "gpt", "value": "Is there tooling for that? Not yet. So we're missing some basic features right now. And so we're continuing to drop out new features on a weekly basis. But at the fullness of time, give us a year and a half, maybe two years. Is it an automatable process?" }, { "from": "human", "value": "So when we're ready, it will be very automatable, yes. Is it automatable? Is it possible to automate, in the general case, the Python to Mojo conversion?" }, { "from": "gpt", "value": "Yeah, well, you're saying it's possible. Yeah, well, you're saying it's possible. Well, so, and this is why, I mean, among other reasons, why we use tabs, right? So first of all, by being a superset, it's like C versus C++. Can you move C code to C++? Yes. Yeah, right. And you can move C code to C++. And then you can adopt classes. You can adopt templates. You can adopt other references or whatever C++ features you want after you move C code to C++. You can't use templates in C. And so if you leave it to C, fine, you can't use the cool features, but it still works. And C and C++ code work together. And so that's the analogy. Now, here, there's not a Python is bad and then Mojo is good. Mojo just gives you superpowers. And so if you want to stay with Python, that's cool. But the tooling should be actually very beautiful and simple because we're doing the hard work of defining" } ]

[ { "from": "human", "value": "a superset. Yes, right. So you're right. So there's several things to say there, but also the conversion tooling should probably" }, { "from": "gpt", "value": "give you hints as to how you can improve the code. And then exactly once you're in the new world, then you can build all kinds of cool tools to say, hey, should you adopt this feature? And we haven't built those tools yet, but I fully expect those tools will exist. And then you can like, quote unquote, modernize your code or however you want to look at it. So I mean, one of the things that I think is really interesting about Mojo is that there have been a lot of projects to improve Python over the years. Everything from getting Python to run on the Java virtual machine, PyPy, which is a JIT compiler. There's tons of these projects out there that have been working on improving Python in various ways. They fall into one of two camps. So PyPy is a great example of a camp that is trying to be compatible with Python. Even there, not really. It doesn't work with all the C packages and stuff like that. But they're trying to be compatible with Python. There's also another category of these things where they're saying, well, Python is too complicated. And I'm going to cheat on the edges. And integers in Python can be an arbitrary size integer. If you care about it going fast in a register in a computer, that's really annoying. And so you can choose to pass on that. You can say, well, people don't really use big integers that often. Therefore, I'm going to just not do it, and it will be fine. Not a Python superset. Or you can do the hard thing and say, OK, this is Python. You can't be a superset of Python without being a superset of Python. And that's a really hard technical problem. But it's, in my opinion, worth it. And it's worth it because it's not about any one package. It's about this ecosystem. It's about what Python means for the world. And it also means we don't want to repeat the Python 2 to Python 3 transition. We want people to be able to adopt this stuff quickly." }, { "from": "human", "value": "And so by doing that work, we can help lift people. Yeah, it's a really interesting technical philosophical challenge of really making a language a superset of another language." }, { "from": "gpt", "value": "It's breaking my brain a little bit. Well, it paints you in the corners. So again, I'm very happy with Python. So all joking aside, I think that the indentation thing is not the actual important part of the problem. But the fact that Python has amazing dynamic metaprogramming features, and they translate to beautiful static metaprogramming features, I think is profound. I think that's huge. And so Python, I've talked with Guido about this. It was not designed to do what we're doing. That was not the reason they built it this way. But because they really cared and they were very thoughtful about how they designed the language, it scales very elegantly in this space. But if you look at other languages, for example, C and C++, if you're building a superset, you get stuck with the design decisions of the subset. And so C++ is way more complicated because of C in the legacy than it would have been if they would have theoretically designed a from scratch thing. And there's lots of people right now that are trying to make C++ better. And re-syntax C++ is going to be great. We'll just change all the syntax. But if you do that, now suddenly you" }, { "from": "human", "value": "have zero packages, you don't have compatibility. So what are the, if you could just linger on that, what are the biggest challenges of keeping that superset status? What are the things you're struggling with?" }, { "from": "gpt", "value": "Is it all boiled down to having a big integer? No, I mean, it's a long-tail weird things. So let me give you a war story. So war story in this space is you go way back in time. Project I worked on is called Clang. Clang, what it is is a C++ parser. And when I started working on Clang, it must have been like 2006 or something. It was 2007, 2006 when I first started working on it. It's funny how time flies. I started that project, and I'm like, OK, well, I want to build a C++ parser for LLVM. It's going to be the work, GCC is yucky. This is me in earlier times. It's yucky. It's on principle. It has all these weird features, all these bugs. It's yucky, so I'm going to build a standard compliant C and C++ parser. It's going to be beautiful. It'll be amazing, well engineered, all the cool things that NGR wants to do. And so I started implementing and building it out, and building it out, and then I got to include standardio.h. And all of the headers in the world use all the GCC stuff. So again, come back away from theory back to reality. I was at a fork in the road. I could have built an amazingly beautiful academic thing" }, { "from": "human", "value": "that nobody would ever use." }, { "from": "gpt", "value": "What are the other things like? Clang. Or I could say, well, it's yucky in various ways. All these design mistakes, accents of history, the legacy. At that point, GCC was over 20 years old, which by the way, now LLVM's over 20 years old. So it's funny how time catches up to you. And so you say, OK, well, what is easier? As an engineer, it's actually much easier for me to go implement long tail compatibility, weird features, even if they're distasteful, and just do the hard work, and figure it out, reverse engineer it, understand what it is, write a bunch of test cases, try to understand behavior. It's way easier to do all that work as an engineer than it is to go talk to all C programmers and get argue with them, and try to get them to rewrite their code, by the way. Yeah." } ]

[ { "from": "human", "value": "And so it's funny how engineer it understand. Because that breaks a lot more things." }, { "from": "gpt", "value": "Yeah. A lot more things. Yeah. And you have realities. Nobody actually even understands how the code works, because it was written by the person who quit 10 years ago. And so this software is kind of frustrating that way. But that's how the world works." }, { "from": "human", "value": "Yeah." }, { "from": "gpt", "value": "Unfortunately, it can never be this perfect, beautiful thing. Well, there are occasions in which you get to build, like you invent a new data structure, or something like that. Or there's this beautiful algorithm that just makes you super happy. And I love that moment. But when you're working with people, and you're working with code, and Dusty Deck code bases, and things like this, right? It's not about what's theoretically beautiful. It's about what's practical, what's real, what people will actually use. And I don't meet a lot of people that say," }, { "from": "human", "value": "I want to rewrite all my code, just for the sake of it. By the way, there could be interesting possibilities. And we'll probably talk about it where AI can help rewrite some code. That might be farther out future. But it's a really interesting one, how that could create more, be a tool in the battle against this monster of complexity that you mentioned. You mentioned Guido, the benevolent dictator for life of Python. What does he think about Mojo?" }, { "from": "gpt", "value": "Have you talked to him much about it? I have talked with him about it. He found it very interesting. We actually talked with Guido before it launched. And so he was aware of it before it went public. I have a ton of respect for Guido for a bunch of different reasons. You talk about walrus operator. And Guido is pretty amazing in terms of steering such a huge and diverse community. And driving it forward, and I think Python is what it is, thanks to him. And so to me, it was really important starting to work on Mojo to get his feedback and get his input and get his eyes on this. Now, a lot of what Guido was and is, I think, concerned about is, have we not fragment the community? We don't want to Python 2 to Python 3 thing. That was really painful for everybody involved. And so we spent quite a bit of time talking about that. And some of the tricks I learned from Swift, for example. So in the migration from Swift, we managed to not just convert Objective-C into a slightly prettier Objective-C, which we did. We then converted, not entirely, but almost an entire community to a completely different language. And so there's a bunch of tricks that you learn along the way that are directly relevant to what we do. And so this is where, for example, you leverage C-Python while bringing up the new thing. That approach is, I think, proven and comes from experience. And so Guido was very interested in, OK, cool. I think that Python is really his legacy. It's his baby. I have tons of respect for that. Incidentally, I see Mojo as a member of the Python family. We're not trying to take Python away from Guido and from the Python community. And so to me, it's really important that we're a good member of that community. And so I think that, again, you would have to ask Guido this. But I think that he was very interested in this notion of, cool, Python gets beaten up for being slow community. Maybe there's a path out of that. And if the future is Python, look at the far outside case on this. And I'm not saying this is Guido's perspective. But there's this path of saying, OK, well, Python can suddenly go all the places it's never been able to go before. And that means that Python can go even further and can have even more impact on the world. So in some sense, Mojo could be seen as Python 4.0. I would not say that." }, { "from": "human", "value": "I think that would drive a lot of people really crazy." }, { "from": "gpt", "value": "Because of the PTSD of the 3.0 2.0. I'm willing to annoy people about Emacs versus VIM or about TAF versus Spaces. But not that one. I don't know. That might be a little bit far even for me. My skin may not be that thick." }, { "from": "human", "value": "But not that one. TAF versus Spaces. But not that one. But the point is the step to it being a super set and allowing all of these capabilities, I think, is the evolution of a language. It feels like an evolution of a language. So he's interested by the ideas that you're playing with, but also concerned about the fragmentation. So what are the ideas you've learned? What are you thinking about? How do we avoid fragmenting the community? Where the Pythonistas and the, I don't know what to call the Mojo people. The magicians can coexist happily and share code and basically just have these big code bases that" }, { "from": "gpt", "value": "are using CPython and more and more moving towards Mojo. The magicians, I like it, are moving towards Mojo. Well, so again, these are lessons I learned from Swift. And here we face very similar problems. In Swift, you have Objective-C, super dynamic. They're very different syntax. But you're talking to people who have large-scale code bases. Apple's got the biggest, largest-scale code base of Objective-C code. And so none of the companies, none of the iOS developers, none of the other developers want to rewrite everything all at once, and so you want to be able to adopt things piece at a time. And so a thing that I found that worked very well in the Swift community was saying, OK, cool. And this is when Swift was very young. As you say, OK, you have a million line of code with Objective-C app. Don't rewrite it all. But when you implement a new feature, go implement that new class using Swift. And so now this, it turns out, is a very wonderful thing for an app developer, but it's a huge challenge for the compiler team and the systems people that are implementing this. And this comes back to, what is this trade-off between doing the hard thing that enables scale versus doing the theoretically pure and ideal thing? And so Swift adopted and built a lot of different machinery to deeply integrate with the Objective-C runtime. And we're doing the same thing with Python. Now, what happened in the case of Swift is that Swift as a language got more and more and more mature over time. And incidentally, Mojo is a much simpler language than Swift in many ways, and so I think that Mojo will develop way faster than Swift for a variety of reasons. But as the language gets more mature, in parallel with that, you have new people starting new projects. And so when the language is mature and somebody's starting a new project, that's when they say, OK, cool. I'm not dealing with a million lines of code. I'll just start and use the new thing for my whole stack. Now, the problem is, again, you come back to where communities and where people that work together, you build a new subsystem or a new feature, a new thing, in Swift, or you build a new thing in Mojo, then you want to end up being used on the other side. And so then you need to work on integration back the other way. And so it's not just Mojo talking to Python. It's also Python talking to Mojo. And so what I would love to see, and I don't want to see this next month, but what I want to see over the course of time is I would love to see people that are building these packages, like NumPy, or TensorFlow, or these packages that are half Python, half C++. And if you say, OK, cool, I want to get out of this Python C++ world into a unified world, and so I can move to Mojo, but I can't give up all my Python clients. Because these libraries get used by everybody, and they're not all going to switch all at once, and maybe never. Well, so the way we should do that is we should bend Python interfaces to the Mojo types. And that's what we did in Swift, and it worked great. I mean, it was a huge implementation challenge for the compiler people, but there's only a dozen of those compiler people, and there are millions of users. And so it's a very expensive, capital-intensive, skill-set intensive problem. But once you solve that problem, it really helps adoption, and it really helps the community progressively adopt technologies. And so I think that this approach will work quite well with the Python and the Mojo world." } ]

[ { "from": "human", "value": "So for a package, port it to Mojo, and then create a Python interface. So how do you just link on these packages, NumPy, PyTorch, TensorFlow? How do they play nicely together? So is Mojo supposed to be? Let's talk about the machine learning ones. Is Mojo kind of vision to replace PyTorch and TensorFlow to incorporate it?" }, { "from": "gpt", "value": "What's the relationship in this? All right, so take a step back. So I wear many hats. So you're angling it on the Mojo side. Mojo's a programming language, and so it" }, { "from": "human", "value": "can help solve the C, C++, Python view that's happening. The FireMojo got me. I'm sorry. We should be talking in modular." }, { "from": "gpt", "value": "Yes, yes, yes. OK, so the FireMojo is amazing. I love it. It's a big deal. The other side of this is the FireMojo is in service of solving some big AI problems. And so the big AI problems are, again, this fragmentation, this hardware nightmare, this explosion of new potential, but it's not getting felt by the industry. And so when you look at how does the modular engine help TensorFlow and PyTorch, it's not replacing them. In fact, when I talk to people, again, they don't like to rewrite all their code, you have people that are using a bunch of PyTorch a bunch of TensorFlow. They have models that they've been building over the course of many years. And when I talk to them, there's a few exceptions, but, generally, they don't want to rewrite all their code. And so what we're doing is we're saying, OK, well, you don't have to rewrite all your code. What happens is, the modular engine goes in there and goes underneath TensorFlow and PyTorch. It's fully compatible and just provides better performance, better predictability, better tooling. It's a better experience that helps lift TensorFlow and PyTorch and make them even better. I love Python, I love TensorFlow, I love PyTorch, right? This is about making the world better" }, { "from": "human", "value": "because we need AI to go further. But if I have a process that trains a model and I have a process that performs inference on that model and I have the model itself, what should I do with that in the long arc of history in terms of if I use PyTorch to train it, should I rewrite stuff in Mojo?" }, { "from": "gpt", "value": "Would that, if I care about performance? Oh, so I mean, again, it depends. So if you care about performance, then writing in a Mojo is gonna be way better than writing in Python. But if you look at LLM companies, for example, you look at OpenAI rumored and you look at many of the other folks that are working on many of these, many of these LLMs and other like innovative machine learning models, on the one hand, they're innovating in the data collection and the model billions of parameters and the model architecture and the RLEHF and the like all the cool things that people are talking about. But on the other hand, they're spending a lot of time writing CUDA kernels. And so you say, wait a second, how much faster could all this progress go if they were not having to hand write all these CUDA kernels? And so there are a few technologies that are out there and people have been working on this problem for a while and they're trying to solve subsets to the problem, again, kind of fragmenting the space. And so what Mojo provides for these kinds of companies is the ability to say, cool, I can have a unifying theory. Right, and again, better together, the unifying theory, the two-world problem or the three-world problem or the n-world problem, like this is the thing that is slowing people down. And so as we help solve this problem, I think it will be very helpful" }, { "from": "human", "value": "for making this whole cycle go faster. So obviously we've talked about the transition from Objective-C to Swift, have designed this programming language. And you've also talked quite a bit about the use of Swift for machine learning context. Why have you decided to move away from maybe an intense focus on Swift for the machine learning context versus sort of designing a new programming language" }, { "from": "gpt", "value": "that happens to be a super-service? You're saying this is an irrational set of life choices I make, or what?" }, { "from": "human", "value": "Did you go to the desert and did you meditate on it? Okay, all right. No, it was bold and needed. And I think, I mean, it's just bold. And sometimes to take those leaps" }, { "from": "gpt", "value": "is a difficult leap to take. Yeah, well, so, okay. I mean, I think there's a couple of different things. So, actually I left Apple back in 2017, like January of 2017. So it's been a number of years that I left Apple. And the reason I left Apple was to do AI. Okay, so, and again, I won't comment on Apple and AI, but at the time, right, I wanted to get into and understand and understand the technology, understand the applications, the workloads, and so I was like, okay, I'm gonna go dive deep into applied AI and then the technology underneath it, right?" }, { "from": "human", "value": "I found myself at Google." }, { "from": "gpt", "value": "And that was like when TPUs were waking up. Exactly, and so I found myself at Google and Jeff Dean, who's a rock star, as you know, right? And in 2017, TensorFlow is like really taking off and doing incredible things. And I was attracted to Google to help them with the TPUs, right, and TPUs are an innovative hardware accelerator platform, have now, I mean, I think proven massive scale and like done incredible things, right? And so one of the things that this led into is a bunch of different projects, which I'll skip over, right? One of which was this Swift for TensorFlow project, right? And so that project was a research project. And so the idea of that is say, okay, well, let's look at innovative new programming models where we can get a fast programming language, we can get automatic differentiation into language, let's push the boundaries of these things in a research setting, right? Now that project, I think lasted two, three years. There's some really cool outcomes of that. So one of the things that's really interesting is, I published a talk at an LLVM conference in 2018. Again, this seems like so long ago about graph program abstraction, which is basically the thing that's in PyTorch 2. And so PyTorch 2, with all this Dynamo real thing, it's all about this graph program abstraction thing from Python bytecodes. And so a lot of the research that was done ended up pursuing and going out through the industry and influencing things. And I think it's super exciting and awesome to see that. But the Swift for TensorFlow project itself did not work out super well. And so there's a couple of different problems with that. One of which is that you may have noticed Swift is not Python. There's a few people that write Python code. Yes. And so it turns out that all of ML" } ]

[ { "from": "human", "value": "is pretty happy with Python. It's actually a problem that other programming languages have as well, that they're not Python. We'll probably maybe briefly talk about Julia, who's a very interesting, beautiful programming language," }, { "from": "gpt", "value": "but it's not Python. Exactly, well, and so like if you're saying, I'm gonna solve a machine learning problem where all the programmers are Python programmers, and you say, the first thing you have to do is switch to a different language. Well, your new thing may be good or bad or whatever, but if it's a new thing, the adoption barrier is massive. It's still possible. Still possible, yeah, absolutely. The world changes and evolves, and there's definitely room for new and good ideas, but it just makes it so much harder. And so lesson learned, Swift is not Python, and people are not always in search of learning a new thing for the sake of learning a new thing, and if you wanna be compatible with all the world's code, turns out, still possible. Meet the world where it is, right? Second thing is that lesson learned is that Swift as a very fast and efficient language, kind of like Mojo, but a different take on it still, really worked well with eager mode, and so eager mode is something that PyTorch does, and it proved out really well, and it enables really expressive and dynamic and easy to debug programming. TensorFlow at the time was not set up for that," }, { "from": "human", "value": "let's say." }, { "from": "gpt", "value": "The timing is also important in this world. Yeah, yeah, and TensorFlow is a good thing, and it has many, many strengths, but you could say Swift for TensorFlow is a good idea except for the Swift and except for the TensorFlow part." }, { "from": "human", "value": "Swift because it's not Python and TensorFlow" }, { "from": "gpt", "value": "because it's not eager mode at the time. It wasn't set up for eager mode at the time. as 1.0. Exactly. And so one of the things about that is in the context of it being a research project, I'm very happy with the fact that we built a lot of really cool technology, we learned a lot of things. I think the ideas went on to have influence in other systems like PyTorch, few people use that right here, right? And so I think that's super cool. And for me personally, I learned so much from it, right? And I think a lot of the engineers that worked on it also learned a tremendous amount. And so, you know, I think that that's just really exciting to see and, you know, I'm sorry that the project didn't work out. I wish it did, of course, right? But, you know, it's a research project and so you're there to learn from it." }, { "from": "human", "value": "Well, it's interesting to think about as 1.0. Exactly. About the evolution of programming as we come up with these whole new set of algorithms in machine learning and artificial intelligence and what's going to win out. Because it could be a new programming language. Yeah. And I mean, I just mentioned Julia. I think there's a lot of ideas behind Julia that Mojo shares. What are your thoughts about Julia in general?" }, { "from": "gpt", "value": "So I will have to say that when we launched Mojo, one of the biggest things I didn't predict was the response from the Julia community. And so I was not, I mean, let me take a step back. I've known the Julia folks for a really long time. They were an adopter of LLVM a long time ago. They've been pushing state-of-the-art in a bunch of different ways. Julia is a really cool system. I had always thought of Julia as being mostly a scientific computing focused environment. And I thought that was its focus. I neglected to understand that one of their missions is to help make Python work end to end. And so I think that was my air for not understanding that. And so I could have been maybe more sensitive to that. But there's major differences between what Mojo is doing and what Julia is doing. So as you say, Julia is not Python. And so one of the things that a lot of the Julia people came out and said is like, okay, well, if we put a ton of more energy and a ton more money or engineering or whatever into Julia, maybe that would be better than starting Mojo. I mean, maybe that's true, but it still wouldn't make Julia into Python. So if you've worked backwards from the goal of, let's build something for Python programmers without requiring them to relearn syntax, then Julia just isn't there, right? I mean, that's a different thing, right? And so if you anchor on, I love Julia and I want Julia to go further, then you can look at it from a different lens. But the lens we were coming at was, hey, everybody is using Python. Python isn't, syntax isn't broken. Let's take what's great about Python and make it even better. And so it was just a different starting point. So I think Julia is a great language. The community is a lovely community. They're doing really cool stuff," }, { "from": "human", "value": "but it's just a slightly different angle. But it does seem that Python is quite sticky. Is there some philosophical almost thing you could say about why Python by many measures seems to be the most popular programming language" }, { "from": "gpt", "value": "in the world? Well, I can tell you things I love about it. Maybe that's one way to answer the question, right? So huge package ecosystem, super lightweight and easy to integrate. It has very low startup time, right? So it was startup time, you mean like learning curve or what? Yeah, so if you look at certain other languages, you say like go and it just takes like Java, for example, takes a long time to JIT compile all the things and then the VM starts up and the garbage closure kicks in and then it revs its engines and then it can cloud through a lot of internet stuff or whatever, right? Python is like scripting, like it just goes, right? Python has very low compile time. Like, so you're not sitting there waiting. Python integrates into notebooks in a very elegant way that makes exploration super interactive and it's awesome, right? Python is also, it's like almost the glue of computing because it has such a simple object representation, a lot of things plug into it. That dynamic metaprogramming thing we were talking about also enables really expressive and beautiful APIs, right? So there's lots of reasons that you can look at technical things that Python has done and say like, okay, wow, this is actually a pretty amazing thing and any one of those you can neglect. People will all just talk about indentation and ignore like the fundamental things, but then you also look at the community side, right?" } ]

[ { "from": "human", "value": "So Python owns machine learning. So what's startup time, you mean like learning curve" }, { "from": "gpt", "value": "or what? Machine learning is pretty big? And it's growing." }, { "from": "human", "value": "And it's growing and important, right? And there's a reputation of prestige to machine learning to where like, if you're a new programmer, you're thinking about like which programming, which do I use? Well, I should probably care about machine learning. Therefore, let me try Python." }, { "from": "gpt", "value": "And builds and builds and builds and builds. And builds and builds and builds. Go back before that. My kids learn Python. Not because I'm telling them to learn Python. What are they going against you or what? Alright, they also learn Scratch. And things like this too, but it's because Python is taught everywhere, because it's easy to learn." }, { "from": "human", "value": "And because it's pervasive. I learned\"; back to my day. learn Java and C++, well, uphill both directions. But yes, I guess Python is the main language" }, { "from": "gpt", "value": "of teaching software engineering in schools now. Yeah, well, if you look at this, there's these growth cycles, right? If you look at what causes things to become popular and then gain in popularity, there's reinforcing feedback loops and things like this. And I think Python has done, again, the whole community has done a really good job of building those growth loops and help propel the ecosystem. And I think that, again, you look at what you can get done with just a few lines of code." }, { "from": "human", "value": "It's amazing. So this kind of self-building loop, it's interesting to understand because when you look at Mojo, what it stands for, some of the features, it seems sort of clear that this is a good direction for programming languages to evolve in the machine learning community. But it's still not obvious that it will because of this, whatever the engine of popularity, of virality, is there something you could speak to?" }, { "from": "gpt", "value": "How do you get people to switch? Yeah, well, I mean, I think that the viral growth loop is to switch people to Unicode. I think the Unicode file extensions are what I'm betting on. I think that's gonna be the thing." }, { "from": "human", "value": "Yeah, tell the kids that you can use the fire emoji" }, { "from": "gpt", "value": "and they'd be like, what? Exactly. Exactly. Well, in all seriousness, I think there's really, I'll give you two opposite answers. One is, I hope if it's useful, if it solves problems, and if people care about those problems being solved, they'll adopt the tech. That's kind of the simple answer. And when you're looking to get tech adopted, the question is, is it solving an important problem people need solved? And is the adoption cost low enough that they're willing to make the switch and cut over and do the pain upfront so that they can actually do it? And so hopefully Mojo will be that for a bunch of people and people building these hybrid packages are suffering. It's really painful. And so I think that we have a good shot of helping people. But the other side is, it's okay if people don't use Mojo. It's not my job to say, everybody should do this. I'm not saying Python is bad. I hope Python, CPython, all these implementations, because Python ecosystem is not just CPython. It's also a bunch of different implementations with different trade-offs. And this ecosystem is really powerful and exciting, as are other programming languages. It's not like TypeScript or something is gonna go away. And so there's not a winner-take-all thing. And so I hope that Mojo is exciting and useful to people," }, { "from": "human", "value": "but if it's not, that's also fine. But I also wonder what the use case for why you should try Mojo would be. So practically speaking, it seems like, so there's entertainment. There's a dopamine hit of saying, holy shit, this is 10 times faster. This little piece of code is 10 times faster in Mojo. Out of the box before he gets to 35,000. Exactly. I mean, just even that, I mean, that's the dopamine hit that every programmer sort of dreams of is the optimization. It's also the drug that can pull you in and have you waste way too much of your life optimizing and over-optimizing, right? But so what do you see that would be like comedy? It's very hard to predict, of course, but if you look 10 years from now, Mojo's super successful. What do you think would be the thing where people try and then use it regularly" }, { "from": "gpt", "value": "and it kind of grows and grows and grows and grows and grows? Well, let's say you talk about dopamine hit. And so again, humans are not one thing. And some people love rewriting their code and learning new things and throwing themselves in the deep end and trying out a new thing. In my experience, most people don't. Like they're too busy, they have other things going on. By number, most people don't like this. I want to rewrite all my code. But even those people, the too busy people, the people that don't actually care about the language, just care about getting stuff done, those people do like learning new things, right? And so you talk about the dopamine rush of 10X faster. Wow, that's cool, I want to do that again. Well, it's also like, here's the thing I've heard about in a different domain. And I don't have to rewrite all my code, I can learn a new trick, right? Well, that's called growth, you know? And so one thing that I think is cool about Mojo, and again, this will take a little bit of time for, for example, the blog posts and the books and like all that kind of stuff to develop and the language needs to get further along. But what we're doing, you talk about types. Like you can say, look, you can start with the world you already know and you can progressively learn new things and adopt them where it makes sense. If you never do that, that's cool, you're not a bad person. If you get really excited about it and want to go all the way in the deep end and rewrite everything and like whatever, that's cool, right? But I think the middle path is actually the more likely one where it's, you know, you come out with a new idea and you discover, wow, that makes my code way simpler, way more beautiful, way faster, way whatever. And I think that's what people like. Now, if you fast forward and you said like 10 years up, right, I can give you a very different answer on that, which is, I mean, if you go back and look at what computers look like 20 years ago, every 18 months, they got faster for free, right? 2x faster every 18 months, it was like clockwork, it was free, right? You go back 10 years ago and we entered in this world where suddenly we had multi-core CPUs and we had GPUs. And if you squint and turn your head, what a GPU is, it's just a many core, very simple CPU thing kind of, right? And so, and 10 years ago, it was CPUs and GPUs and graphics. Today, we have CPUs, GPUs, graphics and AI, because it's so important, because the compute is so demanding, because of the smart cameras and the watches and all the different places that AI needs to work in our lives, it's caused this explosion of hardware. And so part of my thesis, part of my belief of where computing goes, if you look out 10 years from now, is it's not gonna get simpler. Physics isn't going back to where we came from, it's only gonna get weirder from here on out, right? And so to me, the exciting part about what we're building is it's about building that universal platform, which the world can continue to get weird, because again, I don't think it's avoidable, it's physics. But we can help lift people scale, do things with it, and they don't have to rewrite their code every time a new device comes out. And I think that's pretty cool. And so if Mojo can help with that problem, then I think that it will be hopefully quite interesting and quite useful to a wide range of people, because there's so much potential and maybe analog computers will become a thing or something, right? And we need to be able to get into a mode where we can move this programming model forward, but do so in a way where we're lifting people and growing them instead of forcing them" } ]

[ { "from": "human", "value": "to rewrite all their code and exploding them. Do you think there'll be a few major libraries that go Mojo first?" }, { "from": "gpt", "value": "Well, so I mean, the modular engines all Mojo. So again, come back to, we're not building Mojo because it's fun, we're building Mojo" }, { "from": "human", "value": "because we had to dissolve these accelerators. That's the origin story." }, { "from": "gpt", "value": "But I mean, ones that are currently in Python. Yeah, so I think that a number of these projects will. And so one of the things, again, this is just my best guess, like each of the package maintainers also has, I'm sure plenty of other things going on. People like really don't like rewriting code just for the sake of rewriting code. But sometimes like people are excited about like adopting a new idea. Yeah. It turns out that while rewriting code is generally not people's first thing, it turns out that redesigning something while you rewrite it and using a rewrite as an excuse to redesign can lead to the 2.0 of your thing that's way better than the 1.0, right? And so I have no idea, I can't predict that. But there's a lot of these places where again, if you have a package that is half C and half Python, right, you just solve the pain, make it easier to move things faster, make it easier to debug and evolve your tech, adopting Mojo kind of makes sense to start with. And then it gives you this opportunity" }, { "from": "human", "value": "to rethink these things. Yeah. So the two big gains are that there's a performance gain and then there's the portability" }, { "from": "gpt", "value": "to all kinds of different devices. And there's safety, right? So you talk about real types. I mean, not saying this is for everybody, but that's actually a pretty big thing, right? Yeah, types are. And so there's a bunch of different aspects of what value Mojo provides. And so, I mean, it's funny for me, like I've been working on these kinds of technologies and tools for too many years now, but you look at Swift, right? And we talked about Swift for TensorFlow, but Swift as a programming language, right? Swift's now 13 years old from when I started it. So, because I started in 2010, if I remember. And so that project, and I was involved with it for 12 years or something, right? That project has gone through its own really interesting story, right? And it's a mature, successful, used by millions of people system, right? Certainly not dead yet, right? But also going through that story arc, I learned a tremendous amount about building languages, about building compilers, about working with community and things like this. And so that experience, like I'm helping channel and bring directly into Mojo. And other systems, same thing. Like apparently I like building, building and iterating and evolving things. And so you look at this LLVM thing that I worked on 20 years ago, you look at MLIR, right? And so a lot of the lessons learned in LLVM got fed into MLIR. And I think that MLIR is a way better system than LLVM was. And Swift is a really good system and it's amazing. But I hope that Mojo will take the next step forward" }, { "from": "human", "value": "in terms of design, right? Yeah, types are, in terms of running Mojo, people can play with it. What's a Mojo playground? Yeah. And from the interface perspective, and from the hardware perspective," }, { "from": "gpt", "value": "what's this incredible thing running on? Yeah, so right now, so here we are two weeks after launch. We decided that, okay, we have this incredible set of technology that we think might be good, but we have not given it to lots of people yet. And so we were very conservative and said, let's put it in a workbook so that if it crashes, we can do something about it. We can monitor and track that, right? And so again, things are still super early, but we're having like one person a minute sign up" }, { "from": "human", "value": "with over 70,000 people two weeks in, it's kinda crazy. So you can sign up to Mojo Playground" }, { "from": "gpt", "value": "and you can use it in the cloud, in your browser, in your browser. And so what that's running on is that's running on cloud VMs. And so you share a machine with a bunch of other people, but it turns out there's a bunch of them now because there's a lot of people. And so what you're doing is you're getting free compute and you're getting to play with this thing in kind of a limited, controlled way so that we can make sure that it doesn't totally crash and be embarrassing. So, not a lot of the feedback we've gotten is people want to download it locally. So, we're working on that right now. And so that's..." }, { "from": "human", "value": "So that's the goal to be able to download" }, { "from": "gpt", "value": "locally into it? Yeah, that's what everybody expects and so we're working on that right now. And so we just want to make sure that we do it right. And I think this is one of the lessons I learned from Swift also, by the way, is that when we launched Swift, feels like forever ago, it was 2014. It was super exciting. I and we, the team, had worked on Swift for a number of years in secrecy. Four years into this development, roughly, of working on this thing, at that point, about 250 people at Apple knew about it. It was secret. Apple's good at secrecy, and it was a secret project. We launched this at WWDC, a bunch of hoopla on excitement, and said, developers, you're going to be able to develop and submit apps to the app store in three months. Several interesting things happened. First of all, we learned that, hey, it had a lot of bugs. It was not actually production quality, and it was extremely stressful in terms of trying to get it working for a bunch of people. What happened was we went from zero to, I don't know how many developers Apple had at the time, but a lot of developers overnight, and they ran into a lot of bugs, and it was really embarrassing, and it was very stressful for everybody involved. It was also very exciting, because everybody was excited about that. The other thing I learned is that when that happened, roughly every software engineer who did not know about the project at Apple, their head exploded when it was launched, because they didn't know it was coming. They're like, wait, what is this? I signed up to work for Apple because I love objective C. Why is there a new thing? What that meant, practically, is that the push from launch to, first of all, the fall, but then to 2.0 and 3.0 and all the way forward was super painful for the engineering team and myself. It was very stressful. The developer community was very grumpy about it because they're like, okay, well, wait a second. You're changing and breaking my code, and we have to to fix the bugs, and it was just a lot of tension and friction on all sides. There's a lot of technical debt in the compiler because we have to run really fast and you have to go implement the thing and unblock the use case and do the thing. You know it's not right, but you never have time to go back and do it. I'm very proud of the Swift team because they've come, they're heading. I mean, we, but they came so far and made so much progress over this time since launch. It's pretty incredible, and Swift is a very, very good thing," } ]

[ { "from": "human", "value": "but I just don't want to do that again. I'll iterate more through the development process." }, { "from": "gpt", "value": "Yeah, and so what we're doing is we're not launching it when it's, hopefully, 0.9 with no testers. We're launching it and saying it's 0.1, right? And so we're setting expectations of saying, okay, well, don't use this for production, right? If you're interested in what we're doing, we'll do it in an open way, and we can do it together, but don't use it in production yet. We'll get there, but let's do it the right way. I'm also saying, we're not in a race. The thing that I want to do is build the world's best thing. Yeah, right because if you do it right and it lifts the industry it doesn't matter if it takes an extra two months Yeah, like two months is worth waiting. And so doing it right and not being Overwhelmed with technical debt and things like this is like again war wounds Lessons learned whatever you want to say I think is absolutely the right thing to do even though right now people are very frustrated that you know You can't download it or the it doesn't have feature X or something like this" }, { "from": "human", "value": "and so what have you learned in the in a little bit of time since it's been Released into the wild that people have been complaining about future X or Y or Z What have they been complaining about whether they have been excited about like, yeah almost like detailed things versus a big day" }, { "from": "gpt", "value": "I think everyone would be very excited about the big vision. Yeah. Yeah. Well, so, I mean I've been very pleased In fact, I mean, we've been massively overwhelmed with response which is a good problem to have. It's kind of like a success disaster in a sense, right? And so, I mean, if you go back in time, when we started Modular, which was just not yet a year and a half ago. So it's still a pretty new company, new team, small but very good team of people. Like we started with extreme conviction that there's a set of problems that we need to solve. And if we solve it, then people will be interested in what we're doing, right? But again, you're building in basically secret, right? You're trying to figure it out. The creation is a messy process. You're having to go through different paths and understand what you want to do and how to explain it. Often when you're doing disruptive and new kinds of things, just knowing how to explain it is super difficult, right? And so when we launched, we hoped people would be excited. But I'm an optimist, but I also don't want to get ahead of myself. And so when people found out about Mojo, I think their heads exploded a little bit, right? And here's, I think, a pretty credible team that has built some languages and some tools before. And so they have some lessons learned and are tackling some of the deep problems in the Python ecosystem and giving it the love and attention that it should be getting. And I think people got very excited about that. And so if you look at that, I mean, I think people are excited about ownership and taking a step beyond Rust, right? And there's people that are very excited about that. And there's people that are excited about, you know, just, like, I made Game of Life go 400 times faster. Right, and things like that. And that's really cool. There are people that're really excited about, OK. I really hate writing stuff in C++. Save me. Like systems in your, they're, like, stepping up like, yes. So that's me, by the way. Also, I don't really want to stop writing C++." }, { "from": "human", "value": "stop writing C++, but the. Like systems in your, they're like stepping up like. Yeah. Yes. I get third person excitement when people tweet, here I made this code, game of life or whatever." }, { "from": "gpt", "value": "It's faster and you're like. Yeah, and also like, I would also say that, let me cast blame out to people who deserve it. Sure. These terrible people who convinced me to do some of this. Yes. Jeremy Howard. Yes." }, { "from": "human", "value": "That guy. Well, he's been pushing for this kind of thing. He's won this for years. He's wanted this for a long. He's won this for years. For people who don't know a Jeremy Howard. He's like one of the most legit people in the machine learning community. He has a grassroots. He really teaches. He's an incredible educator, he's an incredible teacher, but also legit in terms of a machine learning engineer himself. He's been running the fast.ai and looking I think for exactly what." }, { "from": "gpt", "value": "Exactly. He's been pushing for more years. Yeah, he's a long, long time. He's one of those three years. And so, fuck. Thank you. Thank you. Exactly. And so, I mean, the first time, so I met Jeremy pretty early on. But the first time I sat up and I'm like, this guy is ridiculous, is when I was at Google and we were bringing up TPUs and we had a whole team of people. And there was this competition called Dawn Bench of who can train ImageNet fastest, right? And Jeremy and one of his researchers crushed Google by not through sheer force of the amazing amount of compute and the number of TPUs and stuff like that, that he just decided that progressive image resizing was the right way to train the model. And if you're an epoch faster and make the whole thing go vroom, right? And I'm like, this guy is incredible. And so you can say, anyways, come back to, you know, where's Mojo coming from? Let's finally listen to Jeremy." } ]

[ { "from": "human", "value": "It's all his fault. It's all his fault. Well, there's a kind of very refreshing, pragmatic view that he has about machine learning that I don't know if it's like this mix of desire for efficiency, but ultimately grounded in desire to make machine learning more accessible to a lot of people. I don't know what that is. I guess that's coupled with efficiency and performance, but it's not just obsessed about" }, { "from": "gpt", "value": "performance. Well, so a lot of AI and AI research ends up being that it has to go fast enough to get scale. So a lot of people don't actually care about performance, particularly on the research side, until it allows them to have more, a bigger data set, right? And so suddenly now you care about distributed compute and like all these exotic HPC, like you don't actually want to know about that. You just want to be able to do more experiments faster and do so with bigger data sets, right? And so Jeremy has been really pushing the limits. And one of the things I'll say about Jeremy, and there's many things I could say about Jeremy because I'm a fanboy of his, but he fits in his head. And Jeremy actually takes the time where many people don't to really dive deep into why is the beta parameter of the atom optimizer equal to this, right? And he'll go survey and understand what are all the activation functions in the trade offs and why is it that everybody that does this model pick that thing. So the why, not just trying different values, like really what is going on here? Right. And so as a consequence of that, like he's always, again, he makes time, but he spends time to understand things at a depth that a lot of people don't. And as you say, he then brings it and teaches people and his mission is to help lift his website says, making AI uncool again. Like it's about, like, forget about the hype. It's actually practical and useful. Let's teach people how to do this, right? Now the problem Jeremy struggled with is that he's pushing the envelope, right? Research isn't about doing the thing that is staying on the happy path or the well-paved road, right? And so a lot of the systems today have been these really fragile fragmented things or special case in this happy path." }, { "from": "human", "value": "And if you fall off the happy path, you get eaten by an alligator. So what about, uh, so Python has this giant ecosystem of packages, uh, and there's a package repository. Do you have ideas of how to do that? Well, for Mojo, yeah, how to do a repository packages." }, { "from": "gpt", "value": "Yeah. Well, so that's another really interesting problem that I knew about, but I didn't understand how big of a problem it was. Uh, Python packaging, a lot of people have very big pain points and a lot of scars with Python packaging. Oh, you mean, uh, so there's several things to say building and distributing and several things to say. Building and distributing and managing dependencies and versioning and all this stuff." }, { "from": "human", "value": "So from the, so from the perspective of if you want to create your own package." }, { "from": "gpt", "value": "Yes. Yes. And then, or you want to build on top of a bunch of other people's packages and then they get updated and things like this. Now I'm not an expert in this, so I don't know the answer. I think this is one of the reasons why it's great that we work as a team and there's other really good and smart people involved. Um, the, uh, but one of my, one of the things I've heard from smart people who've done a lot of this is that the packaging becomes a huge disaster when you get the Python and C together. And so if you have this problem where you have code split between Python and C now, not only do you have to package the C code, you have to build the C code. C doesn't have a package manager, right? C doesn't have a dependency versioning management system, right? And so I'm not experiencing the state of the art and, uh, all the different Python package managers, but my understanding is that's a massive part of the problem. And I think Mojo solves that part of the problem directly heads on. Now, one of the things I think we'll do with the community, and this isn't, again, we're not solving all the world's problems at once. We have to be kind of focused to start with is that I think that we will have an opportunity to reevaluate packaging, right? And so I think that we can come back and say, okay, well, given the new tools and technologies and the cool things we have that we've built up, because we have not just syntax, we have an entirely new compiler stack that works in a new way. Maybe there's other innovations we can bring together and maybe we can help solve that" }, { "from": "human", "value": "problem community. So almost a tangent to that question from the user perspective of packages, it was always surprising to me that it was not easier to sort of explore and find packages that, you know, with pip install, it feels, it's an incredible ecosystem, it's just interesting that it wasn't made, it's still, I think, not made easier to discover packages to do," }, { "from": "gpt", "value": "like search and discovery, as YouTube calls it. Yeah. Well, I mean, it's kind of funny because this is one of the challenges of these like intentionally decentralized communities. And so I don't know what the right answer is for Python, I mean, there are many people that didn't, I mean, I don't even know the right answer from Mojo. So there are many people that would have much more informed opinions than I do. But it's interesting if you look at this, right, open source communities, you know, there's Git, Git is a fully decentralized, anybody can do it any way they want, but then there's GitHub, right, and GitHub centralized, commercial in that case, right, thing, really help pull together and help solve some of the discovery problems and help build a more consistent community." } ]

[ { "from": "human", "value": "And so maybe there's opportunities for something like a GitHub for the community. Yeah. Although even GitHub, I might be wrong on this, but the search and discovery for GitHub is not that great. Like I still use Google search." }, { "from": "gpt", "value": "Yeah, well, I mean, yeah, well, I mean, maybe that's because GitHub doesn't want to replace Google search. Right. And I think there is room for specialized solutions to specific problems. Sure. I don't know. I don't know the right answer for GitHub either. I just think they can go figure that out." }, { "from": "human", "value": "Right. But the point is to have an interface that's usable, that's accessible to people of all different skill levels." }, { "from": "gpt", "value": "Well, well, and again, like what are the benefits of standards, right? Standards allow you to build these next level up ecosystem, the next level up infrastructure or next level up things. And so again, come back to hate complexity, C, C plus Python is complicated. It makes everything more difficult to deal with. It makes it difficult to port, move code around, work with, all these things get more complicated. And so, I mean, I'm not an expert, but maybe Mojo can help a little bit by helping reduce the amount of C in this ecosystem and make it therefore scale better." }, { "from": "human", "value": "So, I mean, you kind of pack out of packages that are hybrid in nature would be a natural fit to move to Mojo. Which is a lot of them, by the way. Yeah. A lot of them, especially that are doing some interesting stuff, computationalized. Let me ask you about some features. Yeah. So, we talked about, obviously, indentation, that it's a typed language or optionally typed." }, { "from": "gpt", "value": "Is that the right way to say it? Which is a lot of them. Yeah. So, either optionally or progressively. Progressively. I think, so people have very strong opinions on the right word to use. Yeah." }, { "from": "human", "value": "I don't know. I look forward progressively. Forward to your letters. So, there's the var versus let, but let is for constants, var is an optional. Yeah, var makes it mutable, so you can reassign. Yeah. Okay. Then there's function overloading. Oh, okay. Yeah. I mean, there's a lot of source of happiness for me, but function overloading, that's, I guess, is that for performance or is that, why does Python not have function overloading?" }, { "from": "gpt", "value": "Oh, okay. Yeah. So, I can speculate. So, Python is a dynamic language. The way it works is that Python and Objective C are actually very similar worlds if you ignore syntax. And so, Objective C is straight line derived from Smalltalk, a really venerable, interesting language that much of the world has forgotten about, but the people that remember it love it generally. And the way that Smalltalk works is that every object has a dictionary in it, and the dictionary maps from the name of a function or the name of a value within an object to its implementation. And so, the way you call a method in Objective C is you say, go look up the way I call foo as I go, look up foo. I get a pointer to the function back, and then I call it. That's how Python works, right? And so, now, the problem with that is that the dictionary within a Python object, all the keys are strings, and it's a dictionary, so you can only have one entry per name. Do you think it's as simple as that? I think it's as simple as that. and so now why do they never fix this? Like why do they not change it to not be a dictionary? Why do they not change, do other things? Well you don't really have to in Python because it's dynamic and so you can say I get into the function, now if I got past an integer, do some dynamic test for it if it's a string, go do another thing. There's another additional challenge which is even if you did support overloading, you're saying, okay well here's a version of a function for integers and a function for strings. So even if you could put it in that dictionary, you'd have to have the caller do the dispatch. And so every time you call the function, you'd have to say is it an integer, is it a string? And so you'd have to figure out where to do that test. And so in a dynamic language, overloading is something you don't have to have. So, but now you get into a type language. And in Python, if you subscript with an integer then you get typically one element out of a collection. subscript with a range, you get a different thing out. And so often in typed languages, you'll want to be able to express the fact that, cool, I have different behavior depending on what I actually pass into this thing. If you can model that, it can make it safer and more predictable and faster and all these" }, { "from": "human", "value": "things. You think it's as simple as that? It somehow feels safer, yes, but it also feels empowering. Make it in terms of clarity." }, { "from": "gpt", "value": "You don't have to design whole different functions. Make it in terms of clarity. Yeah, well, this is also one of the challenges with the existing Python typing systems is that in practice, like you take subscript, like in practice, a lot of these functions, they don't have one signature. They actually have different behavior in different cases. And so this is why it's difficult to retrofit this into existing Python code and make it play well with typing." }, { "from": "human", "value": "You kind of have to design for that. Okay, so there's an interesting distinction that people that program Python might be interested in is def versus fn. So it's two different ways to define a function. And fn is a stricter version of def." }, { "from": "gpt", "value": "What's the coolness that comes from the strictness? So here you get into what is the trade-off with the superset? Yes. Okay, so superset, you have to or you really want to be compatible. If you're doing a superset, you've decided compatibility with existing code is the important thing, even if some of the decisions they made were maybe not what you choose. So that means you put a lot of time in compatibility, and it means that you get locked into decisions of the past, even if they may not have been a good thing. Now, systems programmers typically like to control things. And they want to make sure that, not in all cases, of course, and even systems programmers are not one thing. But often, you want predictability. And so one of the things that Python has, for example, as you know, is that if you define a variable, you just say x equals 4. I have a variable named x. Now I say some long name equals 17, print out some long name, oops, but I typoed it. Well, the compiler, the Python compiler doesn't know, in all cases, what you're defining and what you're using. And did you typo the use of it or the definition? And so for people coming from typed languages, again, I'm not saying they're right or wrong, but that drives them crazy because they want the compiler to tell them, you typoed the name of this thing, right? And so what FN does is it turns on, as you say, it's a strict mode. And so it says, okay, well, you have to actually intentionally declare your variables before you use them. That gives you more predictability, more error checking and things like this, but you don't have to use it. And this is a way that Mojo is both compatible because defs work the same way that defs have always worked. But it provides a new alternative that gives you more control and allows certain kinds" } ]

[ { "from": "human", "value": "of people that have a different philosophy to be able to express that and get that." }, { "from": "gpt", "value": "But usually, if you're writing Mojo code from scratch, you'll be using FN. It depends, again, it depends on your mentality, right? It's not that def is Python and FN is Mojo, Mojo has both and it loves both, right? And it really depends on... Time is just straight. Yeah, exactly. Are you playing around and scripting something out? Is it a one-off throwaway script? Cool." }, { "from": "human", "value": "Yeah, Python is great at that. I will still be using FN, but yeah." }, { "from": "gpt", "value": "I love strictness. Okay, well, so... Control. Power. You also like suffering, right? Yes. Go hand in hand. How many pull-ups? Yes." }, { "from": "human", "value": "Go hand in hand. I've lost count at this point. Exactly." }, { "from": "gpt", "value": "So, yeah, exactly. Exactly. And that's cool, I love you for that. And I love other people who like strict things, right? But I don't want to say that that's the right thing because Python's also very beautiful for hacking around and doing stuff and research and these other cases where you may not want that." }, { "from": "human", "value": "See, I just... Yeah. I feel like... Maybe I'm wrong on that, but it feels like strictness leads to faster debugging. So, in terms of going from even on a small project from zero to completion, it just... I guess it depends how many bugs you generate usually." }, { "from": "gpt", "value": "Yeah. You know, it's, again, lessons learned in looking at the ecosystem. It's really... I mean, I think if you study some of these languages over time, like the Ruby community, for example. Now, Ruby is a pretty well-developed, pretty established community, but along their path, they really invested in unit testing. So, I think that the Ruby community has really pushed forward the state-of-the-art of testing because they didn't have a type system that caught a lot of bugs in compile time. Right? And so, you can have the best of both worlds. You can have good testing and good types and things like this. but I thought that it was really interesting to see how certain challenges get solved. And in Python, for example, the interactive notebook kind of experiences and stuff like this are really amazing. And if you typo something, it doesn't matter, it just tells you, that's fine, right? And so I think that the trade-offs are very different if you're building a large-scale production system" }, { "from": "human", "value": "versus you're building and exploring in a notebook. The speaking of control, the hilarious thing, if you look at code I write just for myself for fun, it's like littered with asserts everywhere. Okay, okay. It's a kind of." }, { "from": "gpt", "value": "Yeah, you'd like types." }, { "from": "human", "value": "It's basically saying in a dictatorial way," }, { "from": "gpt", "value": "this should be true now, otherwise everything stops. Well, and that is the sign, again, I love you, man, but that is a sign of somebody who likes control. Yeah. And so, yes, I think that you'll like, I think I'll like Mojo." }, { "from": "human", "value": "Yeah, I think I'll like Mojo. Therapy session, yes, I definitely will. Speaking of asserts, exceptions are called errors." }, { "from": "gpt", "value": "Why is it called errors? So, I mean, we use the same, we're the same as Python, but we implement in a very different way. And so, if you look at other languages, like we'll pick on C++, our favorite, right? C++ has this thing called zero cost exception handling. Okay, and this is, in my opinion, something to learn lessons from. It's a nice polite way of saying it. And so, zero cost exception handling, the way it works is that it's called zero cost because if you don't throw an exception, there's supposed to be no overhead for the non-error code. And so it takes the error path out of the common path. It does this by making throwing an error extremely expensive. And so if you actually throw an error with a C++ compiler using exceptions, it has to go look up in tables on the side and do all this stuff. And so throwing an error could be like 10,000 times more expensive than referring for a function, right? Also, it's called zero cost exceptions, but it's not zero cost by any stretch of the imagination because it massively blows out your code, your binary. It also adds a whole bunch of different paths because of destructors and other things like that that exist in C++. And it reduces the number of optimizations. It adds like all these effects. And so this thing that was called zero cost exceptions, it really ain't, okay? Now, if you fast forward to newer languages and this includes Swift and Rust and Go and now Mojo, well, and Python's a little bit different because it's interpreted. And so like, it's got a little bit of a different thing going on, but if you look at compiled languages, many newer languages say, okay, well, let's not do that zero cost exception handling thing. Let's actually treat and throwing an error the same as returning a variant, returning either the normal result or an error. Now, programmers generally don't want to deal with all the typing machinery and like pushing around a variant. And so you use all the syntax that Python gives us, for example, try and catch, you know, functions that raise and things like this. You can put a raises decorator on your functions, stuff like this, and if you want to control that, and then the language can provide syntax for it. But under the hood, the way the computer executes it, throwing an error is basically as fast as returning something." }, { "from": "human", "value": "It's a nice polite way of saying it, nice polite. Oh, interesting. So it's exactly the same way." }, { "from": "gpt", "value": "It's from a compiled perspective. And so this is actually, I mean, it's a fairly nerdy thing, right? Which is why I love it. But this has a huge impact on the way you design your APIs, right? So in C++, huge communities turn off exceptions because the cost is just so high, right? And so the zero cost cost is so high, right? And so that means you can't actually use exceptions in many libraries, right? And even for the people that do use it, well, okay, how and when do you want to pay the cost? If I try to open a file, should I throw an error? Well, what if I'm probing around looking for something, right, and I'm looking it up in many different paths? Well, if it's really slow to do that, maybe I'll add another function that doesn't throw an error or turns an error code instead. And I have two different versions of the same thing. And so it causes you to fork your APIs. And so, you know, one of the things I learned from Apple and I so love is the art of API design is actually really profound. I think this is something that Python's also done a pretty good job at in terms of building out this large scale package ecosystem. It's about having standards and things like this. And so, you know, we wouldn't want to enter a mode where, you know, there's this theoretical feature that exists in language, but people don't use it in practice. Now, I'll also say one of the other really cool things about this implementation approach is that it can run on GPUs and it can run on accelerators and things like this. And that standard zero cost exception thing would never work on an accelerator. And so this is also part of how Mojo can scale all the way down to like little embedded systems" } ]

[ { "from": "human", "value": "and to running on GPUs and things like that. Can you actually say about the, maybe, is there some high level way to describe the challenge of exceptions and how they work in code during compilation?" }, { "from": "gpt", "value": "So it's just this idea of percolating up a thing, an error. Yeah, yeah, so the way to think about it is think about a function that doesn't return anything, just as a simple case, right? And so you have function one calls function two calls function three calls function four along that call stack that are tri-blocks. And so if you have function one calls function two, function two has a tri-block and then within it, it calls function three, right? Well, what happens to function three throws? Well actually, start simpler, what happens if it returns? Well for returns, is it supposed to go back out and continue executing and then fall off the bottom of the tri-block and keep going and it all's good. If the function throws, you're supposed to exit the current function and then get into the accept clause. And then do whatever codes there and then keep following on and going on. And so the way that a compiler like Mojo works is that the call to that function, which happens in the accept block, calls the function. And then, instead of returning nothing, it actually returns a variant between nothing and an error. And so if you return normally, fall off the bottom and do a return, you return nothing. And if you throw an error, you return the variant that is I'm an error. So when you get to the call, you say, okay, cool. I called a function. Hey, I know locally I'm in a tri-block, right? And so I, I call the function and then I check to see what it returns." }, { "from": "human", "value": "Aha, if it's that error thing, jump to the accept block." }, { "from": "gpt", "value": "And that's all done for you behind the scenes. Exactly. And so the competitors, all this for you. And I mean, one of the things, if you dig into how this stuff works in Python, it gets a little bit more complicated because you have finally blocks, which now you need to go into, do some stuff, and then those can also throw and return. Wait, what? Yeah, in like the stuff matters for compatibility. Um, like there's really, there's, there's with clauses. And so with clauses are kind of like finally blocks with some special stuff going on." }, { "from": "human", "value": "And so there's nesting in general, nesting of anything, nesting of functions should be illegal. It just feels like it adds a level of complexity." }, { "from": "gpt", "value": "Lex, I'm merely an implementer. And so this is again, one of the, one of the, one of the trade-offs you get when you decide to build a super set is you get to implement a full fidelity implementation of the thing that you decided is good. And so, yeah, I mean, we can, we can complain about the reality of" }, { "from": "human", "value": "the world and shake our fist, but it always feels like you shouldn't be allowed to do that, like to declare functions and certain functions inside functions." }, { "from": "gpt", "value": "Oh, wait, wait, wait, what happened to Lex, the LISP guy?" }, { "from": "human", "value": "No. I understand that, but LISP is what I used to do in college. So now you've grown up. You know, we've all done things in college, we're not part of... Wait, wait a sec. I love LISP! I love..." }, { "from": "gpt", "value": "I love LISP. I love LISP! Yeah. I was going to say, you're afraid of me. You're taking a whole internet away." }, { "from": "human", "value": "You're taking the whole internet away. I love LISP." }, { "from": "gpt", "value": "It worked, it worked as a joke on my head, and to come out, right. So nested functions are joking aside, actually really great. For certain things right and so these are also called closures Closures are pretty cool, and you can pass callbacks. There's a lot of good patterns and so" }, { "from": "human", "value": "so speaking of which I don't think you have" }, { "from": "gpt", "value": "Nested functions implemented yet in mojo. Oh, we don't have lambda syntax, but we do have" }, { "from": "human", "value": "There's a few things on the roadmap they have that it'd be cool to sort of just fly through because it's interesting to see You know how many features there are in a language small and big yep. They have to implement yeah So first of all there's tuple support and that has to do with some very specific aspect of it like the parentheses are not parentheses" }, { "from": "gpt", "value": "That yep, they have to implement. Yeah. Yeah, this is just a totally a syntactic thing" }, { "from": "human", "value": "I think that's the thing I think that's the thing okay. There's but it's cool still" }, { "from": "gpt", "value": "So keyword arguments and functions yeah, so this is where in Python you can say call a function x equals 4 Yeah, and x is the name of the yeah argument. That's a nice sort of documenting self-documenting feature feature Yeah, I mean again. This isn't rocket science to implement. That's just the longer. It's just on the list The bigger features are things like traits So traits are when you want to define abstract so when you get into typed languages You need the ability to write generics and so you want to say I won't write this function and now I want to work on all things that are arithmetic like well, what does arithmetic like mean well arithmetic like is a categorization of a bunch of types and so Again, you can define many different ways and I'm not going to go into ring theory or something But the you know you can say it's arithmetic like if you can add subtract multi divide it for example Right and so what you're saying is you're saying There's a set of traits that apply to a broad variety of types, and so there are all these types of arithmetic-like, all these tensors and floating-point integer, and there's this category of types, and then I can define on an orthogonal axis algorithms that then work against types that have those properties. And so this is a, again, it's a widely known thing. It's been implemented in Swift and Rust and many languages, so it's not Haskell, which is where everybody learns their tricks from, but we need to implement that, and that'll" } ]

[ { "from": "human", "value": "enable a new level of expressivity, a new. So classes. Yeah, classes are a big deal. It's a big deal still to be implemented. Like you said, Lambda syntax, and there's like detailed stuff like whole module import. Yeah, classes are a big deal. Support for top-level code and file scope, and then global variables also, so being able" }, { "from": "gpt", "value": "to have variables outside of a top-level code. And so this comes back to the where Mojo came from and the fact that this is 0.1, right? And so we're building, so modular is building an AI stack, right? And an AI stack has a bunch of problems working with hardware and writing high-performance kernels and doing this kernel fusion thing I was talking about and getting the most out of the hardware. And so we've really prioritized and built Mojo to solve modular's problem, right? Now, our North Star is build out and support all the things, and so we're making incredible progress. By the way, Mojo's only like seven months old." }, { "from": "human", "value": "So that's another interesting thing. I mean, part of the reason I wanted to mention some of these things is like, there's a lot to do, and it's pretty cool how you just kind of, sometimes you take for granted how much there is in a programming language, how many cool features you kind of rely on." }, { "from": "gpt", "value": "And this is kind of a nice reminder when you lay it as a to-do list. Yeah, and so, I mean, but also you look into, it's amazing how much is also there, and you take it for granted that a value, if you define it, it will get destroyed automatically. Like that little feature itself is actually really complicated, given the way the ownership system has to work, and the way that works within Mojo is a huge step forward from what Rust and Swift have done. But can you say that again, when a value, when you define it, gets destroyed automatically? Yeah, so like, say you have a string, right? So you just find a string on the stack, whatever that means, like in your local function, right? And so you say, like, whether it be in a def, and so you just say x equals hello world, right? Well, if your string type requires you to allocate memory, then when it's destroyed, you have to deallocate it. So in Python and in Mojo, you define that with the dell method, right? Where does that get run? Well, it gets run sometime between the last use of the value and the end of the program. Like in this, you now get into garbage collection, you get into like all these long debated, you talk about religions and trade offs and things like this, this is a hugely hotly contested world. If you look at C++, the way this works is that if you define a variable, or a set of variables within a function, they get destroyed in a last in first out order. So it's like nesting, okay? This has a huge problem because if you define you have a big scope, and you define a whole bunch of values at the top, and then you use them, and then you do a whole bunch of code that doesn't use them, they don't get destroyed until the very end of that scope, right? And so this also destroys tail calls, so good functional programming, right? This has a bunch of different impacts on, you know, you talk about reference counting optimizations and things like this, a bunch of very low level things. And so what Mojo does is it has a different approach on that from any language I'm familiar where it destroys them as soon as possible. And by doing that you get better memory use, you get better predictability, you get tail calls that work. You get a bunch of other things, you get better ownership tracking. There's a bunch of these very simple things that are very fundamental, that are already built in there in MOJO today, that are the things that nobody talks about generally but when they don't work right you find out and you have to complain about." }, { "from": "human", "value": "Is it trivial to know what's the soon as possible" }, { "from": "gpt", "value": "to delete a thing that's not gonna be used again? Yeah, I mean its generally trivial, it's after the last use of it. So if you find X as a string and then you have some use of X somewhere in your code." }, { "from": "human", "value": "What doesqt mean within scope? What do you mean within scope this accessible?" }, { "from": "gpt", "value": "Yeah, exactly. So you can only use something within its scope. And so then it doesn't wait until the end of the scope to delete it," }, { "from": "human", "value": "it destroys it after the last use. So there's kind of some very eager machine" }, { "from": "gpt", "value": "that's just sitting there and deleting. Yeah, it's all on the compiler so it's not at runtime, which is also cool. And so yeah, and so what and this is actually non trivial because you have control flow, and so it gets complicated pretty quickly and so like getting this right." }, { "from": "human", "value": "Was not, and so it gets Christina, it's got this like in a lot of places" }, { "from": "gpt", "value": "reason potentially compiler has reason about this and this is where again, it's experience building languages and not getting this, right? So, again you get another chance to do it and you get, basic things like this, right? But it's, it's extremely powerful when you do that. Right. And so there's a bunch of things like that, that kind of combine together. And this comes back to the, you get a chance to do it the right way, do it the right way, and make sure that every brick you put down is really good. So that when you put more bricks on top of it, they stack up to something that's beautiful. Listen, this is where..." } ]

[ { "from": "human", "value": "Well, there's also, like how many design discussions do there have to be about particular details, like implementation of particular small features? Because the features that seem small, I bet some of them might be like really require really big design decisions." }, { "from": "gpt", "value": "Yeah, yeah. Well, so I mean, let me give you another example of this. Python has a feature called async await. So it's, it's a new feature. I mean, in the long arc of history, it's a relatively new feature, right? That allows way more expressive asynchronous programming. Okay. Again, this is, this is a, Python is a beautiful thing, and they did things that are great for mojo for completely different reasons. The reason the async await got added to Python, as far as I know, is because Python doesn't support threads. Okay. And so Python doesn't support threads, but you want to work with networking and other things like that that can block. I mean, Python does support threads, it's just not its strength. And so, and so they added this feature called async await. It's also seen in other languages like Swift and JavaScript and many other places as well. async await and mojo is amazing, because we have a high performance header, gs, compute, runtime underneath the covers, that then allows non-blocking IO so you get full use of your accelerator. That's huge. Turns out, it's actually really an important part of fully utilizing the machine. You talk about design discussions. That took a lot of discussions, right? And it probably will require more iteration. And so my philosophy with mojo is that, you know, we have a small team of really good people that are pushing forward, and they're very good at the extremely deep knowing how the compiler and runtime and like all the low level stuff works together. But they're not perfect. Same thing as the Swift team, right? And this is where one of the reasons we released mojo much earlier is so we can get feedback. And we've already like renamed a keyword, did a community feedback and we use an ampersand and now it's named in out. We're not we're not renaming existing Python keywords because that breaks compatibility, right? We're naming things we're adding and making sure that they are designed well, we get usage experience, we iterate and work with the community because, again, if you scale something really fast and everybody writes all their code and they start using it in production, then it's impossible to change. And so you want to learn from people, you want to iterate and work on that early on. And this is where design discussions, it's actually quite" }, { "from": "human", "value": "important. Could you incorporate an emoji like into the language, into the main" }, { "from": "gpt", "value": "language? Like a good, like a good, like a, like a, do you have a, do you have a" }, { "from": "human", "value": "favorite one? Well, I really like in terms of humor, like raw, full, whatever, rolling on the floor, laughing. So that could be like, what would that be the use case for that? Like an exception, throw an exception of some sort? I don't know. You should totally file feature requests. Or maybe a hard one. It has to" }, { "from": "gpt", "value": "be a hard one. People have told me that I'm insane. So this is, this is, I'm" }, { "from": "human", "value": "liking this. I'm going to, I'm going to use the viral nature of the internet to" }, { "from": "gpt", "value": "actually get this, to get this past. I mean, it's funny, you come back to the flame emoji file extension, right? The, you know, we have the option to use the flame emoji, which just even that concept cause, for example, the people that get" }, { "from": "human", "value": "up to say, now I've seen everything. Yeah. There's something, uh, it kind of, it's reinvigorating. It's like, uh, it's like, Oh, that's possible. That's really cool." }, { "from": "gpt", "value": "That for some reason that makes everything else like, I think the world is ready for this stuff. And so, you know, when we have a package manager, we'll clearly have to innovate by having the compiled package thing be the little box" }, { "from": "human", "value": "with the bow on it. Right. I mean, it has to be done. It has to be done. It has to be done. Is there some stuff on the roadmap that you particularly stressed" }, { "from": "gpt", "value": "about or excited about that you're thinking about a lot? I mean, as a today snapshot, which will be absolutely tomorrow, uh, the lifetime stuff is really exciting. And so lifetimes give you safe references to memory without dangling pointers. And so this has been done in languages like Rust before. And so we have a new approach, which is really cool. I'm very excited about that. That'll be out to the community very soon. Um, the traits feature is really a big deal. And so that's blocking a lot of API design. And so there's that. I think that's really exciting. Um, a lot of it is these kind of table stakes features. Um, one of the things that is, uh, again, also lessons learned with Swift, uh, is that, uh, programmers in general like to add syntactic sugar. And so it's like, Oh, well this annoying thing, like, like in Python, you have to spell unbar, unbar add. Why can't I just use plus deaf plus? Come on. Why can't I just do that? Right. And so trigger a bit of syntactic sugar. It makes sense. It's beautiful. It's obvious. We're trying not to do that. And so, um, for two different reasons, one of which is that again, lesson learned with Swift. Swift has a lot of syntactic sugar, um, which may or may be a good thing. Maybe not. I don't know. But, um, but because it's such an easy and addictive thing to do sugar, like make sure blood get crazy, right? Um, like the community will really dig into that and want to do a lot of that. And I think it's very distracting from building the core abstractions. The second is we want to be a good member of the Python community, right? And so we want to work with the broader Python community and yeah, we're pushing forward a bunch of systems programming features and we need to build them out to understand them. But once we get a long ways forward, I want to make sure that we go back to the Python community and say, okay, let's do some design reviews. Let's actually talk about this stuff. Let's figure out how we want this stuff all to work together. And syntactic sugar just makes all that more complicated. So," } ]

[ { "from": "human", "value": "and, uh, yeah, list comprehensions are yet to be implemented. And my favorite did, I mean, dictionaries." }, { "from": "gpt", "value": "Yeah. There's some, there's some basic, there's some, there's some basic," }, { "from": "human", "value": "0.1, 0.1." }, { "from": "gpt", "value": "Yeah. But nonetheless, it's actually so quite interesting and useful." }, { "from": "human", "value": "As you mentioned, modular is very new. Mojo is very new. It's a relatively small team. Yeah. There's building up this gigantic stack, this incredible stack that's going to perhaps define the future of development of our AI overlords. We just hope it will be useful. As do all of us. So what, uh, what have you learned from this process of building up a team? Maybe one question is, how do you hire great programmers, great people that operate in this compiler hardware, machine learning, software interface design space? Yeah. And maybe are a little bit fluid in what they can do. So, okay. So language" }, { "from": "gpt", "value": "design, yeah, we're just, so, okay. So, so building a company is just as interesting in different ways as building a language, like different skill sets, different things, but super interesting. And I've built a lot of teams, a lot of different places. Um, if you zoom in from the big problem into recruiting, well, so here's our problem. Okay. I'll just, I'll be very straightforward about this. We started modular with a lot of conviction about, we understand the problems. We understand the customer pain points. We need to work backwards from the suffering in the industry. And if we solve those problems, we think it'll be useful for people. But the problem is, is that the people we need to hire, as you say, are all these super specialized people that have jobs at big tech, big tech worlds, right? And, you know, we, I don't think we have, um, product market fit in the way that a normal startup does. And we don't have product market fit challenges because right now everybody's using AI and so many of them are suffering and they want help. And so again, we started with strong conviction. Now, again, you have to hire and recruit the best and the best all have jobs. And so what we've done is we said, okay, well let's build an amazing culture. Start with that. That's usually not something a company starts with. Usually you hire a bunch of people and then people start fighting and it turns into a gigantic mess. And then you try to figure out how to improve your culture later. My co-founder, Tim, in particular, is super passionate about making sure that that's right. And we've spent a lot of time early on to make sure that we can scale." }, { "from": "human", "value": "Can you comment, sorry, before we get to the second, what makes for a good culture?" }, { "from": "gpt", "value": "Yeah. Um, so, I mean, there's many different cultures and I have learned many things from several very unique, almost famously unique cultures. And some of them I learned what to do and some of them I learned what not to do. Okay. And so, um, we want an inclusive culture. Uh, I believe in to do. Like amazing people working together. And so I've seen cultures where you have amazing people and they're fighting each other. I see amazing people and they're told what to do like doubt shalt line up and do what I say it doesn't matter if it's the right thing do it right and neither of these is the and I've seen people that have no direction they're just kind of floating in different places and they want to be amazing they just don't know how and so a lot of it starts with have a clear vision right and so we have a clear vision of what we're doing and so I kind of grew up at Apple in my engineering life right and so a lot of the Apple DNA rubbed off on me my co-founder Tim also is like a strong product guy and so what we learned is you know I decided Apple that you don't work from building cool technology you don't work from like come up with a cool product and think about the features you'll have in the big check boxes and stuff like this because if you go talk to customers they don't actually care about your product they don't care about your technology what they care about is their problems right and if your product can help solve their problems well hey they might be interested in that right and so if you speak to them about their problems if you understand you have compassion you understand what people are working with, then you can work backwards to building an amazing product. So the vision starts by defining the problem. And then you can work backwards in solving technology. And at Apple, I think pretty famously said that there's 100 no's for every yes. I would refine that to say that there's 100 not yes for every yes. But famously, if you go back to the iPhone, for example, iPhone 1, many people laughed at it because it didn't have 3G, it didn't have copy and paste. And then a year later, okay, finally, it has 3G, but it still doesn't have copy and paste. It's a joke. Nobody will ever use this product, blah, blah, blah, blah, blah, blah, blah. Right? Well, year three, it had copy and paste and people stop talking about it. And so being laser focused and having conviction, understanding what the core problems are and giving the team the space to be able to build the right tech is really important. And also, I mean, you come back to recruiting. You have to pay well, right? So we have to pay industry leading salaries and have good benefits and things like this. That's a big piece. We're a remote first company. And so we have to... So remote first has a very strong set of pros and cons. On the one hand, you can hire people from wherever they are. And you can attract amazing talent, even if they live in strange places or unusual places. On the other hand, you have time zones. On the other hand, you have everybody on the internet will fight if they don't understand each other. And so we've had to learn how to have a system where we actually fly people in and we get the whole company together periodically and then we get work groups together and we plan and execute together." } ]

[ { "from": "human", "value": "And there's like an intimacy to the in-person brainstorming that I guess you lose, but maybe you don't, maybe if you get to know each other well and you trust each other, maybe you can do that." }, { "from": "gpt", "value": "Yeah. Well, so when the pandemic first hit, I mean, I'm curious about your experience too. The first thing I missed was having whiteboards, right? Those design discussions where like I can high intensity work through things, get things done, work through the problem of the day, understand where you're on, figure out and solve the problem and move forward. But we figured out ways to work around that now with all these screen sharing and other things like that that we do. The thing I miss now is sitting down at a lunch table with the team, the spontaneous things, like the coffee bar things and the bumping into each other and getting to know people outside of the transactional solve a problem over Zoom." }, { "from": "human", "value": "And I think there's just a lot of stuff that I'm not an expert at this. I don't know who is, hopefully there's some people, but there's stuff that somehow is missing on Zoom. Even with the whiteboard, if you look at that, if you have a room with one person at the whiteboard and there's like three other people at a table, first of all, there's a social aspect of that where you're just shooting the shit a little bit, almost like..." }, { "from": "gpt", "value": "Yeah, yeah, and some people are just kind of coming in and..." }, { "from": "human", "value": "Yeah, that. But also while it's a breakout discussion that happens for like seconds at a time, maybe an inside joke, it's like this interesting dynamic that happens that Zoom. You're bonding. Yeah, you're bonding, but through that bonding, you get the excitement. There's certain ideas are like complete bullshit and you'll see that in the faces of others that you won't see necessarily on zoom Like something it feels like that should be possible to do" }, { "from": "gpt", "value": "Without being in person. Well, I mean you're bonding. Yeah. Yeah. Well, I mean being in person is a very different thing Yeah, I don't it's worth it, but you can't always do it. And so again, we're still learning We're also learning is like a humanity with this new reality, right? But but what we found is that getting people together whether it be a team or the company or whatever Is it worth the expense because people work together and are happier After that like it just it just like there's a massive period of time where you like go out and things start getting afraid Put people together and then you realize that we're all working together. We see things the same way We work through the disagreement or the misunderstanding we're talking across each other and then you work much better together And so things like that. I think are really quite important It's worth it." }, { "from": "human", "value": "What about people that are kind of specialized in very different aspects of the stack working together?" }, { "from": "gpt", "value": "What are some interesting challenges there? Yeah, well, so I mean, there's lots of interesting people, as you can tell. I'm hard to deal with, too. You're one of the most lovable people. So there's different philosophies in building teams. For me," }, { "from": "human", "value": "You're one of the most lovable people." }, { "from": "gpt", "value": "And so some people say, hire 10x programmers! And that's the only thing that... whatever that means, right? What I believe in is building well-balanced teams. Teams that have people that are different in them. Like if you have all generals and no troops, or all troops and no generals, or you have all people that think in one way and not the other way, what you get is you get a very biased and skewed and weird situation where people end up being unhappy. And so what I like to do is I like to build teams with people where they're not all the same. You know, we do have teams that are focused on like runtime or compiler GPU or Excel or whatever the speciality is But people bring a different take and have a different perspective And I look for people that complement each other and particularly if you look at leadership teams and things like this You don't want everybody thinking the same way. You want people bringing different perspectives and experiences" }, { "from": "human", "value": "And so I think that's really important that's team. But what about building a company's ambitious as modular? So what? What's the questions there?" }, { "from": "gpt", "value": "Interesting questions there. Oh, I mean so many like so one of the things I love about okay So modular is the first company I built from scratch One of the first things that was profound was I'm not cleaning up somebody else's mess Right. And so if you look at that's liberating just something. It's super liberating and and And also many of the projects I've built in the past have not been core to the product of the company That's a liberating to something. It's Swift is not Apple's product Right. M. L. A. R. Is not Google's revenue machine or whatever. It's not it's it's important But it's like working on the accounting software for you know, the the retail giant or something, right? It's it's it's like enabling infrastructure and technology. And so modular the the tech we're building is Here to solve people's problems like it is directly the thing we're giving to people And so this is a really big difference and what it means for me as a leader But also for many of our engineers is they're working on the thing that matters Mm-hmm, and that's actually pretty I mean again for for a compiler people and things like that That's that's usually not the case right, and so that's that's also pretty exciting and in quite nice, but the one of the ways that this Manifest is it makes it easier to make decisions. And so one of the challenges I've had in other worlds is it's like, okay Well community matters Somehow for the goodness of the world like or open source matters theoretically But I don't want to pay for a t-shirt right or some swag like well t-shirts cost ten bucks each You can have a hundred t-shirts for a thousand dollars to a megacorp a thousand dollars is Uncountably can't count that low" } ]

[ { "from": "human", "value": "Right, but justifying it and getting a t-shirt. By the way, if you'd like a t-shirt, why would 100%" }, { "from": "gpt", "value": "Like a t-shirt. Are you joking? Fire emoji t-shirt. So I will treasure this I'll pass it down to my grandchildren. And so, you know, it's very liberating to be able to side I think that last should have a t-shirt Right, and it becomes very simple" }, { "from": "human", "value": "This this this is awesome So I have to ask you about the one of the interesting developments with large language models Is that they're able to generate? code recently really well Yes to a degree that maybe I be I don't know if you understand but I have I I struggle to understand because it it forces me to ask questions about the nature of programming of the nature of thought because the Language models are able to predict the kind of code. I was about to write so well Yeah, that it makes me wonder like how unique my brain is and where the valuable ideas actually come from like how much do I contribute in terms of Ingenuity Innovation to code I write or design and that kind of stuff When you stand on the shoulders of giants that you're really doing anything and what LLMs are Helping you do is they help you stand on the shoulders of giants in your program. There's mistakes They're interesting that you learn from but I just it would love to get your opinion first high level Yeah of what you think about this impact of larger language models when they do program synthesis when they generate code" }, { "from": "gpt", "value": "Yeah, yeah. Well, so I don't know where it all goes Yeah I'm an optimist and I'm a human optimist right I think that things I've seen are that a lot of the LLMs are really good at crushing leak code projects and They can reverse the link list like crazy well, it turns out there's a lot of Instances of that on the internet and it's a pretty stock thing. And so if you want to see Standard questions answered lMS can memorize all the answers and that can be amazing and also they do generalize that from that And so there's good work on that but But I think that if you in my experience building things building something like you talk about mojo, or you talk about these things you talk about Building an applied solution to a problem. It's also about working with people It's about understanding the problem. What is the product? You want to build? What are the use case? What are the customers? You can't just go survey all the customers because they'll tell you that they want a faster horse. Maybe they need a car, right? And so a lot of it comes into, you know, I don't feel like we have to compete with LLMs. I think they'll help automate a ton of the mechanical stuff out of the way. And just like, you know, I think we all try to scale through delegation and things like this. Delegating things to an LLM, I think is an extremely valuable approach that will help us all scale and be more productive. But I think it's a fascinating companion. But I'd say I don't think that that means that we're going to be done with coding. Sure." }, { "from": "human", "value": "But I think it's a fascinating companion, but there's power in it as a companion. And from there, I would love to zoom in on to Mojo a little bit. Do you think about that? Do you think about LLMs generating Mojo code and helping sort of like, when you design" }, { "from": "gpt", "value": "your programming language, it almost seems like, man, it would be nice to sort of a little" }, { "from": "human", "value": "bit, almost as a way to learn how I'm supposed to use this thing for them to be trained on" }, { "from": "gpt", "value": "some of the Mojo code. So I do lead an AI company, so maybe there will be a Mojo LLM at some point. But if your question is like, how do we make a language to be suitable for LLMs? Yeah. I think the cool thing about LLMs is you don't have to. And so if you look at what is English or any of these other terrible languages that we as humans deal with on a continuous basis, they're never designed for machines. And yet, they're the intermediate representation. They're the exchange format that we humans use to get stuff done. Right? And so these programming languages, they're an intermediate representation between the human and the computer or the human and the compiler, roughly." }, { "from": "human", "value": "And so, I think the LLMs will have no problem learning whatever keyword we pick. Maybe the FiEmoji is gonna- Oh, maybe that's gonna break it. Doesn't tokenize? The reverse of that, it will actually enable it because one of the issues I could see with being a super set of Python is there'll be confusion about the gray area." }, { "from": "gpt", "value": "So it'll be mixing stuff. Maybe that's gonna break it. Doesn't tokenize? Well, I'm a human optimist. I'm also an LLM optimist. I think that we'll solve that problem. We'll solve the edgy bit. But you look at that and you say, okay, well, reducing the rote thing, right? Turns out compilers are very particular and they really want things, they really want the indentation to be right. They really want the colon to be there on your else or else it'll complain, right? I mean, compilers can do better at this, but LLMs can totally help solve that problem. And so I'm very happy about the new predictive coding and copilot type features and things like this," }, { "from": "human", "value": "because I think it'll all just make us more productive. It's still messy and fuzzy and uncertain, unpredictable. But is there a future you see, given how big of a leap GPT-4 was, where you start to see something like LLMs" }, { "from": "gpt", "value": "inside a compiler, or no? I mean, you could do that. Yeah, absolutely. I mean, I think that'd be interesting. Is that wise? Well, I mean, it would be very expensive. So compilers run fast and they're very efficient and LLMs are currently very expensive. There's on-device LLMs and there's other things going on. And so maybe there's an answer there. I think that one of the things that I haven't seen enough of is that, so LLMs to me are amazing when you tap into the creative potential of the hallucinations, right? And so if you're building doing creative brainstorming or creative writing or things like that, the hallucinations work in your favor, if you're writing code that has to be correct, because you're going to ship it in production, then maybe that's not actually a feature. And so, I think that there has been research and there has been work on building algebraic reasoning systems and kind of figuring out more things that feel like proofs. And so I think that there could be interesting work in terms of building more reliable scale systems and that could be interesting. But if you chase that rabbit hole down, the question then becomes, how do you express your intent to the machine? And so maybe you want LLM to provide the spec but you have a different kind of net than actually implements the code." } ]

[ { "from": "human", "value": "Is that why? Right, so it's to use the documentation and inspiration versus the actual implementation. Yeah, yeah, potentially. Since a successful modular will be the thing that runs, I say so jokingly our AI overlords, but AI systems that are used across, I know it's a cliche term, but Internet of Things," }, { "from": "gpt", "value": "so across just each. I'll joke and say like the AGI should be written in Mojo." }, { "from": "human", "value": "Yeah, yeah, AGI should be written in Mojo. You're joking, but it's also possible that it's not a joke. That a lot of the ideas behind Mojo seems like the natural set of ideas that would enable at scale training and inference of AI systems. So just have to ask you about the big philosophical question about human civilization. So folks like Eliezer Yatkovsky are really concerned about the threat of AI. Do you think about the good and the bad that can happen" }, { "from": "gpt", "value": "at scale deployment of AI systems? Yeah, well, so I've thought a lot about it and there's a lot of different parts to this problem, everything from job displacement to Skynet, things like this. And so you can zoom into sub parts of this problem. I'm not super optimistic about AGI being solved next year. I don't think that's gonna happen personally." }, { "from": "human", "value": "So you have a kind of Zen-like calm about, is there's a nervousness because the leap of GPT-4 seemed so big. Sure. It's huge. It's like we're almost, there's some kind of transitionary period. You're thinking." }, { "from": "gpt", "value": "Well, so, sure. It's a sectionary period. Well, so, I mean, there's a couple of things going on there. One is I'm sure GPT-5 and 7 and 19 will be also huge leaps. They're also getting much more expensive to run. And so there may be a limiting function in terms of just expense on one hand and train, like that could be a limiter that slows things down. But I think the bigger limiter is outside of like Skynet takes over and I don't spend time thinking about that because if Skynet takes over and kills us all, then I'll be dead. So I don't worry about that. So, I mean, that's just okay. Other things worry about, I'll just focus on. Yeah. I'll focus not worry about that one. But I think that the, the other thing I'd say is that AI moves quickly, but humans move slowly and we adapt slowly. And so what I expect to happen is, just like any technology diffusion, like the promise and then the application takes time to roll out. And so I think that I'm not even too worried about autonomous cars defining away all the taxi drivers. Remember, autonomy was supposed to be solved by 2020. Boy, do I remember that. And so like, I think that on the one hand, we can see amazing progress, but on the other hand we can see that reality is a little bit more complicated and it may take longer to roll out than," }, { "from": "human", "value": "than you might expect. Well, that's in the physical space. I do think in the digital space is the stuff that's built on top of LLMs that runs, the millions of apps that could be built on top of them, and they could be run on millions of devices, millions of types of devices. I just think that the rapid effect" }, { "from": "gpt", "value": "it hasn't, human civilization could be truly transformative to it, we don't even know transformative to it." }, { "from": "human", "value": "Well, and there I think it depends on are you an optimist or a pessimist or masochist. Just to clarify, optimist about human civilization." }, { "from": "gpt", "value": "Me too, me too. And so I look at that as saying, okay, cool, well, yeah, I do. And so some people say, oh my God, is it going to destroy us all, how do we prevent that? I kind of look at it from a, is it going to unlock us all, right? You talk about coding, is it going to make us all, I don't have to do all the repetitive stuff. Well, suddenly that's a very optimistic way to look at it. And you look at what a lot of these technologies have done to improve our lives, and I want" }, { "from": "human", "value": "that to go faster. What do you think the future of programming looks like in the next 10, 20, 30, 50 years? The LLMs and with Mojo, with Modular, like your vision for devices, the hardware to compile" }, { "from": "gpt", "value": "others to the different stacks of software. Yeah. Well, so what I want, I mean, coming back to my arch nemesis, right? It's complexity, right? So again, me being the optimist, if we drive down complexity, we can make these tools, these technologies, these cool hardware widgets accessible to way more people, right? And so what I'd love to see is more personalized experiences, more things, the research getting into production instead of being lost at NeurIPS, right, and these things that impact people's lives by entering products. And so one of the things that I'm a little bit concerned about is right now, the big companies are investing huge amounts of money and are driving the top line of AI capability for really quickly, but if it means that you have to have $100 million to train a model or more, $100 billion, right, well, that's going to make it very concentrated with very few people in the world that can actually do this stuff. I'd much rather see lots of people across the industry be able to participate in uses, right? And you look at this, you know, I mean, a lot of great research has been done in the health world and looking at detecting pathologies and doing radiology with AI and doing all these things. Well, the problem today is that to deploy and build these systems, you have to be an expert in radiology and an expert in AI, and if we can break down the barriers so that more people can use AI techniques, and it's more like programming Python, which roughly everybody can do if they want to, right, then I think that we'll get a lot more practical application of these techniques and a lot more niche-ier, cool, but narrower demands, and I think that's going to be really cool. Do you think we'll have more or less programmers in the world than now? Well, so, I think a lot more programmers, but they may not consider themselves to be programmers. That would be a different name for it. Right, I mean, do you consider somebody that uses, you know, I think that arguably the most popular programming language is Excel, right? Yeah. And so do they consider themselves to be programmers? Maybe not, I mean, some of them make crazy macros and stuff like that, but what, you Steve Jobs, it's the bicycle for the mind that allows you to go faster, right? And so I think that as we look forward, right, what is AI? I look at it as hopefully a new programming paradigm. It's like object-oriented programming, right? If you want to write a cat detector, you don't use for loops. It turns out that's not the right tool for the job, right? And so right now, unfortunately, because, I mean, it's not unfortunate, but it's just kind of where things are, AI is this weird different thing that's not integrated into programming languages and normal tool chains, and all the technology is really weird and doesn't work right, and you have to babysit it, and every time you switch hardware, it's different. It shouldn't be that way. When you change that, when you fix that, suddenly, again, the tools and technologies can be way easier to use. You can start using them for many more things, and so that's why I would be excited about" } ]

[ { "from": "human", "value": "it. What kind of advice could you give to somebody in high school right now or maybe early college who's curious about programming and feeling like the world is changing really quickly here? What kind of stuff to learn? What kind of stuff to work on? Should they finish college? Should they go work at a company? Should they build a thing?" }, { "from": "gpt", "value": "What do you think? What do you think? Well, so, I mean, one of the things I'd say is that you'll be most successful if you work on something you're excited by, and so don't get the book and read the book cover to cover and study and memorize and recite and flashcard and go build something. Go solve a problem. Go build the thing that you wanted to exist. Go build an app, train a model, go build something and actually use it and set a goal for yourself. And if you do that, then there's a success, there's the adrenaline rush, there's the achievement, there's the unlock that I think is where if you keep setting goals and you keep doing things and building things, learning by building is really powerful. In terms of career advice, everybody's different. It's very hard to give generalized advice. I'll speak as a compiler nerd. If everybody's going left, sometimes it's pretty cool to go right. Yeah. And so just because everybody's doing a thing, it doesn't mean you have to do the same thing and follow the herd. In fact, I think that sometimes the most exciting paths through life lead to being curious about things that nobody else actually focuses on, right? And it turns out that understanding deeply parts of the problem that people want to take for granted makes you extremely valuable and specialized in ways that the herd is not. And so again, I mean, there's lots of rooms for specialization, lots of rooms for generalists. There's lots of room for different kinds and parts of the problem, but I think that it's, just because everybody's doing one thing" }, { "from": "human", "value": "doesn't mean you should necessarily do it. And now the herd is using Python, so if you want to be a rebel, go check out Mojo and help Chris and the rest of the world fight the arch nemesis of complexity, because simple is beautiful. There you go. Because you're an incredible person. You've been so kind to me ever since we met. You've been extremely supportive. I'm forever grateful for that. Thank you for being who you are, for being legit, for being kind, for fighting this really interesting problem of how to make AI accessible to a huge number of people," }, { "from": "gpt", "value": "huge number of devices. Yeah, well, so Lex, you're a pretty special person too, right? So I think that one of the funny things about you is that besides being curious and pretty damn smart, you're actually willing to push on things, and I think that you've got an agenda to make the world think," }, { "from": "human", "value": "which I think is a pretty good agenda. It's a pretty good one." }, { "from": "gpt", "value": "Thank you so much for talking to me, Chris." }, { "from": "human", "value": "Yeah, thanks, Lex. Thanks for listening to this conversation with Chris Ladner. To support this podcast, please check out our sponsors in the description. And now, let me leave you with some words from Isaac Asimov. I do not fear computers. I fear the lack of them. Thank you for listening, and hope to see you next time." } ]

[ { "from": "human", "value": "This episode is brought to you by Element, spelled L-M-N-T. It's an electrolyte drink mix that I'm currently drinking that I drink throughout the day. I drink a huge amount of it. My favorite flavor is the watermelon salt flavor. It doesn't mean it will be your favorite flavor, but I'm pretty sure it's gonna be your favorite flavor. For fasting, for low carb diets, for all kinds of diets, really, but certainly for low carbohydrate keto carnivore, you have to get the electrolytes right. They call it the keto flu if you don't get the electrolytes right. You know for doing all kinds of crazy exercise that I do, all of that. You have to get the sodium, the potassium, magnesium, and element. Great job of balancing all of that. Makes it delicious. Makes it really easy to make sure" }, { "from": "gpt", "value": "that you're getting just the water intake right. I enjoy this." }, { "from": "human", "value": "Element doesn't, because it basically makes water taste great and balances out the electrolytes, the hydration, everything for you. I bring it, when I travel, I bring it anywhere I go. I have to have element as part of my life. It makes me happy. It makes me feel like I got my stuff together. Anyway, get a simple pack for free with any purchase. Try it at drinkelement.com slash Lex. This show is also brought to you by NetSuite, an all-in-one cloud business management system. You can manage financials, HR, inventory, e-commerce if you do that kind of thing, and many business-related details. Running a company is really complicated. There's a lot of people involved, a lot of tasks involved, a lot of roles involved. It's not just engineering. It's not just design, idea, strategy, vision, all those kinds of things. It's all of the glue, the thing that makes the thing a cohesive, singular system that works efficiently and flawlessly. And so you have to hire the right people. You have to use the best tools for the job. NetSuite does that in the cloud to manage all kinds of messy business details and make it super easy. You can start now with no payment or interest for six months. Go to netsuite.com slash Lex to access their one-of-a-kind financing program. That's netsuite.com slash Lex. This episode is also brought to you by BetterHelp. It's spelled H-E-L-P, help. They figure out what you need and match you with a licensed professional therapist in under 48 hours. Talking to somebody through all the difficult things that are going through your mind, taking those thoughts, taking those memories, converting them into words and letting those words leave your mouth sounds somehow transforms your own understanding of what those memories, those thoughts mean. It allows you to move past them, to integrate them into a healthier, a deeper understanding of the world around you, your own emotions of your own way of being. It's kind of amazing that talk therapy, that talking, talking with a person that knows what they're doing is a powerful way to do that kind of growth. It's funny and it's amazing when simple things like therapy can have a big profound impact, results on your life. Anyway, one of the big things about BetterHelp is just how easy it is to get started. I think that's a barrier for a lot of people. So it's really important to say this, easy, discreet. It is affordable and it's available everywhere worldwide. Check them out at betterhelp.com slash Lex and save on your first month. That's betterhelp.com slash Lex. This is a Lex Friedman podcast to support it. Please check out our sponsors in the description. And now, dear friends, here's Neil Gershenfeld. You have spent your life working at the boundary between bits and atoms, so the digital and the physical. What have you learned about engineering and about nature reality from working at this divide," }, { "from": "gpt", "value": "trying to bridge this divide? I learned why von Neumann and Turing made fundamental mistakes. I learned the secret of life. Yeah. I learned how to solve many of the world's most important problems, which all sound presumptuous," }, { "from": "human", "value": "but all of those are things I learned at that boundary. Okay, so Turing and von Neumann, let's start there. Some of the most impactful, important humans" }, { "from": "gpt", "value": "who have ever lived in computing, why were they wrong? Were they wrong? So I worked with Andy Gleason, who is Turing's counterpart. So just for background, if anybody doesn't know, Turing is credited with the modern architecture of computing, among many other things. Andy Gleason was his US counterpart. And you might not have heard of Andy Gleason, but you might have heard of the Hilbert problems. And Andy Gleason solved the fifth one. So he was a really notable mathematician. During the war, he was Turing's counterpart. Then von Neumann is credited with the modern architecture of computing, and one of his students was Marvin Minsky. So I could ask Marvin what Johnny was thinking, and I could ask Andy what Alan was thinking. And what came out from that, what I came to appreciate as background, I never understood the difference between computer science and physical science, but Turing's machine that's the foundation of modern computing has a simple physics mistake, which is the head is distinct from the tape. So in the Turing machine, there's a head that programmatically moves and reads and writes the tape. The head is distinct from the tape, which means persistence of information is separate from interaction with information. Then von Neumann wrote deeply and beautifully about many things but not computing. He wrote a horrible memo called the first draft of a report on the EdVAC, which is how you program a very early computer. In it, he essentially roughly took Turing's architecture and built it into a machine. So the legacy of that is the computer somebody's using to watch this is spending much of its effort moving information from storage transistors to processing transistors, even though they have the same computational complexity. So in computer science, when you learn about computing, there's a ridiculous taxonomy of about a hundred different models of computation, but they're all fictions. In physics, a patch of space occupies space, it stores state, it takes time to transit, and you can interact. That is the only model of computation that's physical. Everything else is a fiction. So I really came to appreciate that a few years back when I did a keynote for the annual meeting of the supercomputer industry and then went into the halls and spent time with the supercomputer builders and came to appreciate, oh, let's see, if you're familiar with the movie, The Metropolis, people would frolic upstairs in the gardens and down in the basement, people would move levers. And that's how computing exists today, that we pretend software is not physical, it's separate from hardware, and the whole canon of computer science is based on this fiction that bits aren't constrained by atoms, but all sorts of scaling issues in computing come from that boundary, but all sorts of opportunities come from that boundary. And so you can trace it all the way back to Turing's machine making this mistake between the head and the tape. Van Neumann, he never called it Van Neumann's architecture, he wrote about it in this dreadful memo, and then he wrote beautifully about other things we'll talk about. Now, to end a long answer, Turing and Van Neumann both knew this. So all of the canon of computer scientists credits them for what was never meant to be a computer architecture. Both Turing and Van Neumann ended their life studying exactly how software becomes hardware. So Van Neumann studied self-reproducing automata, how a machine communicates its own construction. Turing studied morphogenesis, how genes give rise to form. They ended their life studying the embodiment of computation, something that's been forgotten by the canon of computing, but developed sort of off to the sides" } ]

[ { "from": "human", "value": "by a really interesting lineage. So there's no distinction between the head and the tape, between the computer and the computation," }, { "from": "gpt", "value": "it is all computation. Right, so I never understood the difference between computer science and physical science. And working at that boundary helped lead to things like my lab was part of doing with a number of interesting collaborators, the first Fester than classical quantum computations. We were part of a collaboration creating the minimal synthetic organism where you design life in a computer. Those both involve domains where you just can't separate hardware from software. The embodiment of computation is embodied" }, { "from": "human", "value": "in these really profound ways. So the first quantum computations, synthetic life, so in the space of biology, the space of physics at the lowest level and the space of biology at the lowest level. So let's talk about CBA, Center of Bits and Atoms. What's the origin story of this MIT," }, { "from": "gpt", "value": "legendary MIT center that you're a part of creating? Creating. In high school, I really wanted to go to vocational school where you learn to weld and fix cars and build houses. And I was told, no, you're smart, you have to sit in a room. And nobody could explain to me why I couldn't go to vocational school. I then worked at Bell Labs, this wonderful place before deregulation, legendary place. And I would get union grievances because I would go into the workshop and try to make something. And they would say, no, you're smart, you have to tell somebody what to do. And it wasn't until MIT, and I'll explain how CBA started, but I could create CBA, that I came to understand this is a mistake that dates back to the Renaissance. So in the Renaissance, the liberal arts emerged and liberal doesn't mean politically liberal, this was the path to liberation, birth of humanism. And so the liberal arts were the trivium, quadrivium, roughly language, natural science. And at that moment, what emerged was this dreadful concept of the illiberal arts. So anything that wasn't the liberal arts was for commercial gain, and was just making stuff and wasn't valid for serious study. And so that's why we're left with learning to well wasn't a subject for serious study, but the means of expression have changed since the Renaissance. So micromachining or embedded coding is every bit expressive as painting a painting or writing a sonnet. So never understanding this difference between computer science and physical science, the path that led me to create CBA with colleagues was I was what's called the junior fellow at Harvard. I was visiting MIT through Marvin because I was interested in the physics of musical instruments. This will be another slight digression. In Cornell, I would study physics, and then I would cross the street and go to the music department where I played the bassoon and I would trim reeds and play the reeds, and they'd be beautiful, but then they'd get soggy. And then I discovered in the basement of the music department at Cornell was David Borden, who you might not have heard of, but is legendary in electronic music because he was really the first electronic musician. So Bob Mogge, who invented Mogge Synthesizers was a physics student at Cornell, like me crossing the street, and eventually he was kicked out and invented electronic music. David Borden was the first musician who created electronic music. So he's legendary for people like Phil Glass and Steve Reich. And so that got me thinking about, I would behave as a scientist in the music department, but not in the physics department, but not in the music department, got me thinking about what's the computational capacity of a musical instrument. And through Marvin, he introduced me to Todd Backover at the Media Lab, who was just about to start a project with Yo-Yo Ma that led to a collaboration to instrument a cello to extract Yo-Yo's data and bring it out into computational environments." }, { "from": "human", "value": "What is the computational capacity of a musical instrument? Because we'll continue on this tangent" }, { "from": "gpt", "value": "and we'll share return to CBA. Yeah, so one part of that is to understand the computing. And if you look at like the finest time scale and length scale you need to model the physics, it's not heroic. A good GPU can do teraflops today. That used to be a national class supercomputer, now it's just a GPU. And that's about if you take the time scales and length scales relevant for the physics, that's about the scale of the physics computing. For Yo-Yo, what was really driving it was he's completely unsentimental about the Strad. It's not that it makes some magical wiggles in the sound wave, it's performance as a controller, how he can manipulate it as an interface device. Interface between one and what exactly? Him and sound. Okay, him and sound. So what it led to was I had started by thinking about ops per second, but Yo-Yo's question was really resolution and bandwidth. It's how fast can you measure what he does and the bandwidth and the resolution of detecting his controls and then mapping them into sounds. And what we found, what he found was if you instrument everything he does and connect it to almost anything, it sounds like Yo-Yo, that the magic is in the control, not in ineffable details in how the wood wiggles. And so with Yo-Yo and Todd, that led to a piece and towards the end I asked Yo-Yo what it would take for him to get rid of his Strad and use our stuff. And his answer was just logistics. It was at that time, our stuff was like a rack of electronics and lots of cables and some grad students to make it work. Once the technology becomes as invisible as the Strad, then sure, absolutely, he would take it. And by the way, as a footnote on the footnote, an accident in the sensing of Yo-Yo's cello led to a hundred million dollar a year auto safety business to control airbags and cars. How did that work? I had to instrument the bow without interfering with it. So I set up local electromagnetic fields where I would detect how those fields interact with the bow he's playing. But we had a problem that his hand, whenever his hand got near these sensing fields, I would start sensing his hand rather than the materials on the bow. And I didn't quite understand what was going on with that interference. So my very first grad student ever, Josh Smith, did a thesis on tomography with electric fields, had a C in 3D with electric fields. Then through Todd and at that point, research scientists in my lab, Joe Paradiso, it led to a collaboration with Penn and Teller who, where we did a magic trick in Las Vegas to contact Houdini. And sort of these fields are sort of like contacting spirits. So we did a magic trick in Las Vegas. And then the crazy thing that happened after that was Phil Rittmuller came running into my lab. He worked with, this became with Honda and NEC, airbags were killing infants in rear-facing child seats. Cars need to distinguish a front-facing adult where you'd save the life versus a bag of groceries where you don't need to fire the airbag versus a rear-facing infant where you would kill it. And so the seat need to, in effect, see in 3D to understand the occupants. And so we took the Penn and Teller magic trick derived from Josh's thesis from Yo-Yo's cello to an auto show and all the car companies said, great, when can we buy it? And so that became Elysses and it was $100 million a year business making sensors. There wasn't a lot of publicity because it was in the car so the car didn't kill you. So they didn't sort of advertise, we have nice sensors so the car doesn't kill you." } ]

[ { "from": "human", "value": "But it became a leading auto safety sensor. And that started from the cello and the question of the computational capacity of the musical instrument." }, { "from": "gpt", "value": "So now to get back to MIT, I was spending a lot of outside time at IBM Research that had gods of the foundations of computing. There's just amazing people there. And I'd always expected to go to IBM to take over a lab, but at the last minute pivoted and came to MIT to take a position in the media lab and start what became the predecessor to CBA. Media lab is well known for Nicholas Negroponte. What's less well known is the role of Jerry Wiesner. So Jerry was MIT's president before that Kennedy science advisor, grand old man of science. At the end of his life, he was frustrated by how knowledge was segregated. And so he wanted to create a department of none of the above, a department for work that didn't fit in departments. And the media lab in a sense was a cover story for him to hide a department. As MIT's president towards the end of his tenure, if he said, I'm gonna make a department for things that don't fit in departments, the departments would have screamed. But everybody was sort of paying attention to Nicholas creating the media lab and Jerry kind of hid in it a department called Media Arts and Sciences. It's really the department of none of the above. And Jerry explaining that and Nicholas then confirming it is really why I pivoted and went to MIT. Because my students who helped create quantum computing or synthetic life get degrees from Media Arts and Sciences, this department of none of the above. So that led to coming to MIT. Yeah. With Todd and Joe Paradiso and my colleague, we started a consortium called Things That Think. And this was around the birth of internet of things and RFID. But then we started doing things like work we can discuss that became the beginnings of quantum computing and cryptography and materials and logic and microfluidics. And those needed a much more significant infrastructure and were much longer research arcs. So with a bigger team of about 20 people, we wrote a proposal to the NSF to assemble one of every tool" }, { "from": "human", "value": "to make anything of any size was roughly the proposal." }, { "from": "gpt", "value": "One of any tool to make anything of any size. Yeah. So they're usually nanometers, micrometers, millimeters, meters are segregated, input and output is segregated. We wanted to look just very literally how digital becomes physical and physical becomes digital. And fortunately, we got NSF on a good day and they funded this facility of one of almost every tool to make anything. And so with a group of core colleagues that included Joe Jacobson, Ike Trang, Scott Manales," }, { "from": "human", "value": "we launched CBA. And so you're talking about nanoscale, microscale, nanostructures, microstructures, microstructures, electron microscopes and focused on beam probes for nanostructures, laser, micro machining and X-ray micro tomography for microstructures, multi-axis machining and 3D printing for macrostructures, just some examples. What are we talking about in terms of scale? How can we build tiny things and big things all in one place?" }, { "from": "gpt", "value": "Yeah, so a well-equipped research lab has the sort of tools we're talking about, but they're segregated in different places. They're typically also run by technicians where you then have an account and a project and you charge. All of these tools are essentially when you don't know what you're doing, not when you do know what you're doing. They're when you need to work across length scales, where we don't, once projects are running in this facility, we don't charge for time. You don't make a formal proposal to schedule and the users really run the tools. And it's for work that's kind of inchoate that needs to span these disciplines and length scales. And so work in the project today, work in CBA today ranges from developing Zeptojoule electronics for the lowest power computing to micromachining diamond to take 10 million RPM bearings for molecular spectroscopy studies up to exploring robots to build" }, { "from": "human", "value": "100 meter structures in space. Okay, can we, the three things you just mentioned, let's start with the biggest. What are some of the biggest stuff you attempted" }, { "from": "gpt", "value": "to explore how to build in a lab? Sure, so viewed from one direction, what we're talking about is a crazy random seeming of almost unrelated projects. But if you rotate 90 degrees, it's really just a core thought over and over again, just very literally how bits and atoms relate, how digital and just going from digital to physical in many different domains, but it's really just the same idea over and over again. So to understand the biggest things, let me go back to bring in now Shannon as well as Van Neumann. Claude Shannon. Yeah, so what is digital? Yeah, the casual obvious answer is digital in one and zero, but that's wrong. There's a much deeper answer, which is Claude Shannon at MIT wrote the best master's thesis ever. In his master's thesis, he invented our modern notion of digital logic. Where it came from was Vannevar Bush was a grand old man at MIT. He created the post-war research establishment that led to the National Science Foundation. And he made an important mistake, which we can talk about. But he also made the differential analyzer, which was the last grade analog computer. So it was a room full of gears and pulleys and the longer it ran, the worse the answer was. And Shannon worked on it as a student and he got so annoyed in his master's thesis, he invented digital logic. But he then went on to Bell Labs and what he did there was communication was beginning to expand. There was more demand for phone lines. And so there's a question about how many phone lines, phone messages you could send down a wire. And you could try to just make it better and better. He asked a question nobody had asked, which is rather than make it better and better, what's the limit to how good it can be? And he proved a couple of things, but one of the main things he proved was a threshold theorem for channel capacity. And so what he showed was my voice to you right now is coming as a wave through sound and the further you get, the worse it sounds. But people watching this are getting it as in from packets of data in a network. When they get, when the computer, they're watching this gets the packet of information, it can detect and correct an error. And what Shannon showed is if the noise in the cable to the people watching this is above a threshold, they're doomed. But if the noise is below a threshold for a linear increase in the energy representing our conversation, the error rate goes down exponentially. Exponentials are fast. There's very few of them in engineering. And the exponential reduction of error below a threshold if you restore state is called a threshold theorem. That's what led to digital. That means unreliable things can work reliably. So Shannon did that for communication. Then Van Neumann was inspired by that and applied it to computation. And he showed how an unreliable computer can operate reliably by using the same threshold property of restoring state. It was then forgotten many years. We had to rediscover it in effect in the quantum computing era when things are very unreliable again. But now to go back to, how does this relate to the biggest things I've made? So in fabrication, MIT invented computer controlled manufacturing in 1952. Jet aircraft were just emerging. There was a limit to turning cranks on a machine, on a milling machine to make parts for jet aircraft. Now, this is a messy story. MIT actually stole computer controlled machining from an inventor who brought it to MIT, wanted to do a joint project with the Air Force and MIT effectively stole it from him. So it's kind of a messy history. But that sounds like the birth of computer controlled machining, 1952. There are a number of inventors of 3D printing. One of the companies spun off from my lab by Max Lebowski's Form Labs, which is now a billion dollar 3D printing company. That's the modern version. But all of that's analog, meaning the information is in the control computer. There's no information in the materials. And so it goes back to Vannevar Bush's analog computer. If you make a mistake in printing or machining, just the mistake accumulates. The real birth of computerized digital manufacturing is four billion years ago. That's the evolutionary age of the ribosome. So the way you're manufactured is there's a code that describes you, the genetic code. It goes to a micromachine, the ribosome, which is this molecular factory that builds the molecules that are you. The key thing to know about that is there are about 20 amino acids that get assembled. And in that machinery, it does everything Shannon and Vannevar taught us. You detect incorrect errors. So if you mix chemicals, the error rate is about a part in 100. When you make elongated protein in the ribosome, it's about a part in 10 to the four. When you replicate DNA, there's an extra level of error correction. It's a part in 10 to the eight. And so in the molecules that make you, you can detect incorrect errors, and you don't need a ruler to make you. The geometry comes from your parts. So now compare a child playing with Lego and a state-of-the-art 3D printer or computerized milling machine. The tower made by a child is more accurate than their motor control because the act of snapping the bricks together gives you a constraint on the joints. You can join bricks made out of dissimilar materials. You don't need a ruler for Lego because the geometry locally gives you the global parts and there's no Lego trash. The parts have enough information to disassemble them." } ]

[ { "from": "human", "value": "Those are exactly the properties of a digital code." }, { "from": "gpt", "value": "The unreliable is made reliable. Yes, absolutely. So what the ribosome figured out four billion years ago is how to embody these digital properties, but not for communication or computation in effect, but for construction. So a number of projects in my lab have been studying the idea of digital materials. And think of a digital material just as Lego bricks. The precise meaning is a discrete set of parts reversibly joined with global geometry determined from local constraints. And so it's digitizing the materials. And so I'm coming back to what are the biggest things I've made. My lab was working with the aerospace industry. So Spirit Era was Boeing's factories. They asked us for how to join composites. When you make a composite airplane, you make these giant wing and fuselage parts. And they asked us for a better way to stick them together because the joints were a place of failure. And what we discovered was instead of making a few big parts, if you make little loops of carbon fiber and you reversibly link them in joints and you do it in a special geometry that balances being under-constrained and over-constrained with just the right degrees of freedom, we set the world record for the highest modulus ultralight material just by in effect making carbon fiber Lego. So lightweight materials are crucial for energy efficiency. This let us make the lightest weight high modulus material. We then showed that with just a few part types, we can tune the material properties. And then you can create really wild robots that instead of having a tool the size of a jumbo jet to make a jumbo jet, you can make little robots that walk on these cellular structures to build the structures where they error correct their position on the structure and they navigate on the structure. And so using all of that with NASA, NASA, we made morphing airplanes, a former student Kenny Chung and Ben Jeanette made a morphing airplane the size of NASA, Langley's biggest wind tunnel. With Toyota, we've made super efficiency race cars. We're right now looking at projects with NASA to build these for things like space telescopes and space habitats, where the ribosome, I who I mentioned a little while back can make an elephant one molecule at a time. Ribosomes are slow. They run at about one molecule a second, but ribosomes make ribosomes. So you have thousands of them, trillions of them, and that makes an elephant. In the same way, these little assembly robots I'm describing can make giant structures at heart because the robot can make the robot. So more recently, to my students, Amira and Miana had a nature communication paper showing how this robot can be made out of the parts it's making so the robots can make the robots. So you build up the capacity of robotic assembly." }, { "from": "human", "value": "It can self-replicate. Can you linger on what that robot looks like? What is a robot that can walk along and do error correction? And what is a robot that can self-replicate from the materials that it's given? What does that look like? What are we talking about?" }, { "from": "gpt", "value": "This is fascinating. Yeah, so. This is fascinating. Yeah, the answer is different at different length scales. So to explain that, in biology, primary structure is the code in the messenger RNA that says what the ribosome should build. Secondary structure are geometrical motifs. They're things like helices or sheets. Tertiary structures are functional elements like electron donors or acceptors. Quaternary structure is things like molecular motors that are moving my mouth or making the synapses work in my brain. So there's that hierarchy of primary, secondary, tertiary, quaternary. Now, what's interesting is if you want to buy electronics today from a vendor, there are hundreds of thousands of types of resistors or capacitors or transistors. Huge inventory. All of biology is just made from this inventory of 20 parts, the amino acids. And by composing them, you can create all of life. And so as part of this digitization of materials, we're in effect trying to create something like amino acids for engineering, creating all of technology from 20 parts. Let's see, as another discretion, I helped start an office for science in Hollywood. And there was a fun thing for the movie The Martian where I did a program with Bill Nye and a few others on how to actually build a civilization on Mars that they described in a way that I like as I was talking about how to go to Mars without luggage. And at heart, it's sort of how to create life in non-living materials. If you think about this primary, secondary, tertiary, quaternary structure, in my lab, we're doing that but on different length scales for different purposes. So we're making micro robots out of like nano bricks and to make the robots to build large scale structures in space, the elements of the robots now are centimeters rather than micrometers. And so the assembly robots for the bigger structures are, there are the cells that make up the structure but then we have functional cells. And so cells that can process and actuate, each cell can like move one degree of freedom or attach or detach or process. Now, those elements I just described, we can make out of the still smaller parts. So eventually there's a hierarchy of the little parts make little robots that make bigger parts of bigger robots up through that hierarchy. In that way you can move up to length scale. Early on I tried to go in a straight line from the bottom to the top and that ended up being a bad idea. Instead, we're kind of doing all of these in parallel and then they're growing together. And so to make the larger scale structures, like there's a lot of hype right now about 3D printing houses where you have a printer the size of the house. We're right now working on using swarms of these table scale robots that walk on the structures to place the parts much more efficiently." } ]

[ { "from": "human", "value": "That's amazing. But you're saying you can't for now go" }, { "from": "gpt", "value": "from the very small to the very large." }, { "from": "human", "value": "That'll come, that'll come in stages. Can we just link around this idea starting from Vinoinman's self replicating automata" }, { "from": "gpt", "value": "that you mentioned? Yeah, it's just a beautiful idea. So that's at the heart of all of this. In the stack I described, so one student, Will Langford made these micro robots out of little parts that then we're using for Miana's bigger robots up through this hierarchy. And it's really realizing this idea of the self reproducing automata. So Van Neumann, when I complained about the Van Neumann architecture, it's not fair to Van Neumann because he never claimed it as his architecture. He really wrote about it in this one fairly dreadful memo that led to all sorts of lawsuits and fights and about the early days of computing. He did beautiful work on reliable computation and unreliable devices. And towards the end of his life, what he studied was how, and I have to say this precisely, how a computation communicates its own construction. So beautiful. So a computation can store a description of how to build itself. But now there's a really hard problem, which is if you have that in your mind, how do you transfer it and wake up a thing that then can contain it? So how do you give birth to a thing that knows how to make itself? And so with Stan Ulam, he invented cellular automata as a way to simulate these. But that was theoretical. Now the work I'm describing in my lab is fundamentally how to realize it, how to realize self reproducing automata. This is something Van Neumann thought very deeply and very beautifully about theoretically. And it's right at this intersection, it's not communication or computation or fabrication, it's right at this intersection where communication and computation meets fabrication. Now, the reason self reproducing automata intellectually is so important cause this is the foundation of life. This is really just understanding the essence of how to life. And in effect, we're trying to create life in non-living material. The reason it's so important technologically is because that's how you scale capacity. That's how you can make an elephant from a ribosome" }, { "from": "human", "value": "cause the assemblers make assemblers. So simple building blocks that inside themselves contain the information how to build more building blocks and between each other construct" }, { "from": "gpt", "value": "arbitrarily complex objects. Yeah, right. Now let me give you the numbers. So let me relate this to, right now we're living in AI mania explosion time. Let me relate that to what we're talking about. A hundred petaflop computer, which is a current generation supercomputer, not quite the biggest ones, does 10 to the 17 ops per second. Your brain does 10 to the 17 ops per second. It has about 10 to the 15 synapses and they run at about a hundred Hertz. So as of a year or two ago, theet performance of a big computer matched a brain. So you could view AI as a breakthrough, but the real story is within about a year or two ago, and I see that the supercomputer has about 10 to the 15 transistors in the processors, 10 to the 15 transistors in the memory, which is the synapses in your brain. So the real breakthrough was the computers match the computational capacity of a brain, and so we'd be sort of derelict if they couldn't do about the same thing. But now the reason I'm mentioning that is the chip fab making the supercomputer is placing about 10 to the 10 transistors a second. While you're digesting your lunch right now, you're placing about 10 to the 18 parts per second. There's an eight order of magnitude difference. So in computational capacity, it's done, we've caught up. But there's eight orders of magnitude difference in the rate at which biology can build versus state of the art manufacturing can build. And that distinction is what we're talking about. That distinction is not analog, but this deep sense of digital fabrication, of embodying codes in construction. So a description doesn't describe a thing," }, { "from": "human", "value": "but the description becomes the thing. So you're saying, I mean, this is one of the cases you're making and this is the third revolution, we've seen the Moore's Law in communication, we've seen the Moore's Law like type of growth in computation and you're anticipating, we're going to see that in digital fabrication. Can you actually, first of all," }, { "from": "gpt", "value": "describe what you mean by this term digital fabrication? So the casual meaning is the computer controls the tool to make something. And that was invented when MIT stole it in 1952. Yeah. There's the deep meaning of what the ribosome does, of a digital description doesn't describe a thing, a digital description becomes the thing. That's the path to the Star Trek replicator and that's the thing that doesn't exist yet. Now, I think the best way to understand what this roadmap looks like is to now bring in fab labs and how they relate to all of this. What are fab labs? So here's a sequence. With colleagues, I accidentally started a network of what's now 2,500 digital fabrication community labs called fab labs right now in 125 countries and they double every year and a half. That's called Lassa's Law after Sherry Lassiter, who I'll explain. So here's the sequence. We started Center for Bits and Atoms to do the kind of research we're talking about. We had all of these machines and then had a problem. It would take a lifetime of classes to learn to use all the machines. So with colleagues who helped start CBA, we began a class modestly called How to Make Almost Anything. And there's no big agenda. It was aimed at a few research students to use the machines and it were completely unprepared for the first time we taught it. We were swamped by every year since hundreds of students try to take the class. It's one of the most oversubscribed classes at MIT. Students would say things like, can you teach this at MIT? It seems too useful. It's just how to work these machines. And the students in the class, I would teach them all the skills to use all these tools and then they would do projects integrating them and they're amazing. So Kelly was a sculptor, no engineering background. Her project was she made a device that saves up screams" } ]

[ { "from": "human", "value": "when you're mad and plays them back later and. Saves up screams when you're mad" }, { "from": "gpt", "value": "and plays in the back later. You scream into this device and it deadens the sound, records it, and then when it's convenient, releases your scream." }, { "from": "human", "value": "So can we just pause on the brilliance of that invention? Creation, the art, I don't know, the brilliance. Who is this that created this?" }, { "from": "gpt", "value": "Kelly Dobson. Going on to do a number of interesting things. Meejin, who's gone on to do a number of interesting things, made a dress instrumented with sensors and spines. And when somebody creepy comes close, it would defend your personal space. They're also very easy. Another project early on was a web browser for parrots, which have the cognitive ability of a young child and lets parrots surf the internet. You know, another was an alarm clock you wrestle with and prove you're awake." }, { "from": "human", "value": "And what connects all of these is." }, { "from": "gpt", "value": "They're also very easy. So MIT made the first real-time computer, the whirlwind. That was transistorized as the TX. The TX was spun off from MIT as the PDP. PDPs were the mini-computers that created the internet. So outside MIT was DEC, Prime, Wang, Data General, the whole mini-computer industry. The whole computing industry was there. And it all failed when computing became personal. Ken Olson, the head of digital, famously said, you don't need a computer at home. There's a little background to that, but DEC completely missed computing became personal. So I mentioned all of that because I was asking how to do digital fabrication, but not really why. The students in this how to make class were showing me that the killer app of digital fabrication" }, { "from": "human", "value": "is personal fabrication." }, { "from": "gpt", "value": "Yeah, how do you jump to the personal fabrication? So Kelly didn't make the screen body because it was for a thesis. She wasn't writing a research paper. It wasn't a business model. It was because she wanted one. It was personal expression, going back to me in vocational school, is personal expression in these new means of expression." }, { "from": "human", "value": "So that's happened every year since. It literally is called, the course is literally called how to make almost anything." }, { "from": "gpt", "value": "A legendary course at MIT, every year thing, every year. And it's grown to multiple labs at MIT with as many people involved in teaching as taking it." }, { "from": "human", "value": "And there's even a Harvard lab for the MIT class. What have you learned about humans colliding with the fab lab about what the capacity of humans" }, { "from": "gpt", "value": "to be creative and to build? I mentioned Marvin, another mentor at MIT, sadly no longer living, is Seymour Pappar. So Pappar studied with Piaget. He came to MIT to get access to the early compu-, Piaget was a pioneer in how kids learn. Pappar came to MIT to get access to the early computers with the goal of letting kids play with them. Piaget helped show kids are like scientists. They learn as scientists and it gets kind of throttled out of them. Seymour wanted to let kids have a broader landscape to play. Seymour's work led with Mitch Resnick to Lego, logo, mindstorms, all of that stuff. As Phab Lab spread and we started creating educational programs for kids in them, Seymour said something really interesting. He made a gesture, he said, it was a thorn in his side that they invented what's called the turtle, a robot kids could, early robot kids could program to connect it to a mainframe computer. Seymour said, the goal was not for the kids to program the robot, it was for the kids to create the robot. And so in that sense, the Phab Labs, which for me were just this accident he described as sort of this fulfillment of the arc of kids learn by experimenting, it was to give them the tools to create, not just assemble things and program things, but actually create. So coming to your question, what I've learned is MIT, a few years back, somebody added up businesses spun off from MIT and it's the world's 10th economy, it falls between India and Russia. And I view that in a way as a bad number because it's only a few thousand people and these aren't uniquely the 4,000 brightest people, it's just a productive environment for them. And what we found is in rural Indian villages in African Shanny towns and Arctic hamlets, I find exactly precisely that profile. So Ling cited a few hours above Tromsø, way above the Arctic circles. It's so far north, the satellite dishes look at the ground, not the sky. Hans Christian in the lab was considered a problem in the local school because they couldn't teach him anything. I showed him a few projects, next time I came back, he was designing and building little robot vehicles. And in South Africa, I mentioned Soshant Govee in this apartheid township, the local technical institute taught kids how to make bricks and fold sheets, it was punitive. But Chapiso in the fab lab was actually doing all the work of my MIT classes. And so over and over, we found precisely the same kind of bright invent of creativity. And historically, the answer was, you're smart, go away. It's sort of like me in vocational school. But in this lab network, what we could then do is, in effect, bring the world to them. Now let's look at the scaling of all of this. So there's one earth, a thousand cities, a million towns, a billion people, a trillion things. There was one whirlwind computer, MIT made the first real-time computer. There were thousands of PDPs, there were millions of hobbyist computers that came from that, billions of personal computers, trillions of internet of things. So now if we look at this fab lab story, 1952 was the NC mill, there are now thousands of fab labs, and the fab lab costs exactly the same cost and complexity of the mini computer. So on the mini computer, it didn't fit in your pocket, it filled the room, but video games, email, word processing, really anything you do, the internet, anything you do with a computer today happened at that era because it got on the scale of a work group, not a corporation. In the same way, fab labs are like the mini computers inventing, how does the world work if anybody can make anything? Then if you look at that scaling, labs today are transitioning from buying a machine to machines making machines. So we're transitioning to, you can go to a fab lab, not to make a project but to make a new machine. So we talked about the deep sense of self-replication, there's a very practical sense of fab lab machines making fab lab machines. And so that's the equivalent of the hobbyist computer era, what it's called the Altair historically. Then the work we spent a while talking about about assemblers and self-assemblers, that's the equivalent of smartphones and internet of things. That's when... So the assemblers are like the smartphone where a smartphone today has the capacity of what used to be a supercomputer in your pocket. And then the smartphone thermostat on your wall has the power of the original PDP computer, not metaphorically, but literally. And now there's trillions of those in the same sense that when we finally merge materials with the machines and the self-assembly, that's like the internet of things stage. But here's the important lesson. If you look at the computing analogy, computing expanded exponentially, but it really didn't fundamentally change. The core things happened in that transition in the mini computer era. So in the same sense, the research now we spent a while talking about is how we get to the replicator. Today, you can do all of that if you close your eyes and view the whole fab lab as a machine. In that room, you can make almost anything, but you need a lot of inputs. Bit by bit, the inputs will go down and the size of the room will go down" } ]

[ { "from": "human", "value": "as we go through each of these stages. So how difficult is it to create a self-replicating assembler, self-replicating machine that builds copies of itself or builds more complicated version of itself, which is kind of the dream towards which you're pushing" }, { "from": "gpt", "value": "in a generic arbitrary sense? I had a student, Nadia Peek with Jonathan Ward, who for me started this idea of how do we use the tools in my lab to make the tools in the lab? Yes. In a very clear sense, they are making self-reproducing machines. So one of the really cool things that's happened is there's a whole network of machine builders around the world. So there's Danielle and now in Germany and Jens in Norway. And each of these people has learned the skills to go into a fab lab and make a machine. And so we've started creating a network of super fabs. So the fab lab can make a machine, but it can't make a number of the precision parts of the machine. So in places like Bhutan or Kerala in the south of India, we've started creating super fab labs that have more advanced tools to make the parts of the machines so that the machines themselves become even cheaper. So that is self-reproducing machines, but you need to feed it things like bearings or microcontrollers, they can't make those parts. But other than that, they're making their own things. And I should note as a footnote, the stack I described of computers controlling machines to machine making machines, to assemblers to self assemblers view that as fab one, two, three, four. So we're transitioning from fab one to fab two and the research in the lab is three and four. At this fab two stage, a big component of this is sustainability in the material feedstocks. So Alicia colleague in Chile is leading a great effort looking at how you take forest products and coffee grounds and seashells and a range of locally available materials and produce the high tech materials that go into the lab. So all of that is machine building today. Then back in the lab, what we can do today is we have robots that can build structures and can assemble more robots and build structures. We have finer resolution robots that can build micromechanical systems. So robots that can build robots that can walk and manipulate. And we're just now, we have a project at the layer below that where there's endless attention today to billion dollar chip fab investments. But a really interesting thing we pass through is today the smallest transistors you can buy as a single transistor just commercially for electronics is actually the size of an early transistor in an integrated circuit. So we're using these machines, making machines, making assemblers to place those parts to not use a billion dollar chip fab to make integrated circuits," }, { "from": "human", "value": "but actually assemble little electronic components. So have a fine enough, precise enough actuators" }, { "from": "gpt", "value": "and manipulators that allow you to place these transistors. Right, that's a research project in my lab called DICE on discrete assembly of integrated electronics. And we're just at the point to really start to take seriously this notion of not having a chip fab make integrated electronics, but having not a 3D printer, but a thing that's a cross between a pick and place make circuit boards in 2D. The 3D printer extrudes in 3D. We're making sort of a micromanipulator that acts like a printer," }, { "from": "human", "value": "but it's placing to build electronics in 3D. But this micromanipulator is distributed. So there's a bunch of them," }, { "from": "gpt", "value": "or is this one centralized thing? So that's why that's a great question. So I have a prize that's almost but not been claimed for the students whose thesis can walk out of the printer. Oh, nice. So you have to print the thesis with the means to exit the printer, and it has to contain its description of the thesis" }, { "from": "human", "value": "that says how to do that. It's a really good, I mean, it's a fun example" }, { "from": "gpt", "value": "of exactly the thing we're talking about. And I've had a few students almost get to that. And so in what I'm describing, there's the stack where we're getting closer, but it's still quite a few years to really go from us. So there's a layer below the transistors where we assemble the base materials that become the transistor. We're now just at the edge of assembling the transistors to make the circuits. We can assemble the micro parts to make the micro robots. We can assemble the bigger robots. And in the coming years," }, { "from": "human", "value": "we'll be patching together all of those scales. So do you see a vision of just endless billions of robots at the different scales, self-assembling," }, { "from": "gpt", "value": "self-replicating and building the more complicated structures? Yes. And the but to the yes, but is, let me clarify two things. One is that immediately raises King Charles' fear of Grey Goo, of runaway mutant self-reproducing things. The reason why there are many things I can tell you to worry about, but that's not one of them, is if you want things to autonomously self-reproduce and take over the world, that means they need to compete with nature on using the resources of nature of water and sunlight. And in light of everything I'm describing, biology knows everything I told you. Every single thing I explain, biology already knows how to do. What I'm describing isn't new for biology, it's new for non-biological systems. So in the digital era, the economic win ended up being centralized, the big platforms in this world of machines that can make machines. I'm asked, for example, what's the killer opportunity? Who's gonna make all the money? Who to invest in? But if the machine can make the machine, it's not a great business to invest in the machine. In the same way that if you can think globally, but produce locally, then the way the technology goes out into society isn't a function of central control, but is fundamentally distributed. Now, that raises an obvious kind of concern, which is, well, doesn't this mean you could make bombs and guns and all of that? The reason that's much less of a problem than you would think is making bombs and guns and all of that is a very well-met market need. Anywhere we go, there's a fine supply chain for weapons. Now, hobbyists have been making guns for ages and guns are available just about anywhere. So you could go into the lab and make a gun. Today, it's not a very good gun and guns are easily available. And so generally, we've run these lab in war zones. What we find is people don't go to them to make weapons, which you can already do anyway. It's an alternative to making weapons. Coming back to your question, I'd say the single most important thing I've learned is the greatest natural resource of the planet is this amazing density of bright and vene of people whose brains are underused. And you could view the social engineering of this lab work is creating the capacity for them. And so in the end, the way this is gonna impact society isn't gonna be command and control. It's how the world uses it. And it's been really gratifying for me to see just how it does." } ]

[ { "from": "human", "value": "Yeah, but what are the different ways the evolution of the exponential scaling of digital fabrication can evolve? So you said, yeah, self-replicating nanobots, right? This is the Grey Goo fear. It's a caricature of a fear, but nevertheless, there's interesting, just like you said, spam and all these kinds of things that came with the scaling of communication and computation. What are the different ways that malevolent actors" }, { "from": "gpt", "value": "will use this technology? Yeah, well, first, let me start with a benevolent story, which is trash is an analog concept. There's no trash in a forest. All the parts get disassembled and reused. Trash means something doesn't have enough information to tell you how to reuse it. It's as simple as there's no trash in a Lego room. When you assemble Lego, the Lego bricks have enough information to disassemble them. So as you go through this fab one, two, three, four story, one of the implications of this transition from printing to assembling. So the real breakthrough technologically isn't additive versus subtractive, which is a subject of a lot of attention and hype. 3D printers are useful. We spun off companies like Formlabs, led by Max for 3D printing. But in a fab lab, it's one of maybe 10 machines. It's used, but it's only part of the machines. The real technological change is when we go from printing and cutting to assembling and disassembling. But that reduces inventories of hundreds of thousands of parts to just having a few parts to make almost anything. It reduces global supply chains to locally sourcing these building blocks. But one of the key implications is it gets rid of technological trash because you can disassemble and reuse the parts, not throw them away. And so initially that's of interest for things at the end of long supply chains, like satellites on orbit. But one of the things coming is eliminating technical trash" }, { "from": "human", "value": "through reuse of the building blocks. So like when you think about 3D printers, you're thinking about addition and subtraction. When you think about the other options available to you in that parameter space, as you call it, it's going to be assembly, disassembly cutting, you said." }, { "from": "gpt", "value": "So the 1952 NC Mill was subtractive. You remove material. And 3D printing additive, and there's a couple of claims to the invention of 3D printing, that's closer to what's called net shape, which is you don't have to cut away the material you don't need. You just put material where you do need it. And so that's the 3D printing revolution. But there are all sorts of limitations on 3D printing to the kinds of materials you can print, the kind of functionality you can print. We're just not going to get to making everything in a cell phone on a single printer. But I do expect to make everything in a cell phone with an assembler. And so instead of printing and cutting technologically, it's this transition to assembling and disassembling. Going back to Shannon and Van Neumann, going back to the ribosome four billion years ago. Now, you come to malevolent. Let me tell you a story about, I was doing a briefing for the National Academy of Sciences group that advises the intelligence communities. And I talked about the kind of research we do. And at the very end, I showed a little video clip of Valentina and Ghana making, a local girl making surface mount electronics in the fab lab. And I showed that to this room full of people. One of the members of the intelligence community got up livid and said, how dare you waste our time showing us a young girl in an African village making surface mount electronics. We're looking at, we need to know about disruptive threats to the future of the United States. And somebody else got up in the room and yelled at him, and you idiot, I can't think of anything more important than this. But for two reasons. One reason was because if we rely on like informational superiority in the battlefield, it means other people could get access to it. But this intelligence person's point, bless him, wasn't that it was getting at the root causes of conflict is if this young girl in an African village could actually master surface mount electronics, it changes some of the most fundamental things about recruitment for terrorism, impact of economic migration, basic assumptions about an economy." }, { "from": "human", "value": "It's just existential for the future of the planet. But we've just lived through a pandemic. I would love to linger on this because the possibilities that are positive are endless, but the possibilities are negative are still nevertheless extremely important. Well, it's both positive and negative. What do you do with a large number of general assemblers?" }, { "from": "gpt", "value": "Yes, with the fab lab, you could roughly make a bio lab then learn biotechnology. Now that's terrifying because making self reproducing gray goo that outcompetes biology, I consider doom because biology knows everything I'm describing and is really good at what it does. In how to grow almost anything, you learn skills in biotechnology" }, { "from": "human", "value": "that let you make serious biological threats. When you combine some of the innovations you see with large language models, some of the innovations you see with alpha fold, so applications of AI for designing biological systems, for writing programs, which you can't let large language models increasingly. There seems to be an interesting dance here of automating the design stage of complex systems using AI, and then that's the bits. Now, the innovations you're talking about, you can leap from the complex systems in the digital space to the printing, to the creation, to the assembly at scale of complex systems in the physical space." }, { "from": "gpt", "value": "Yeah, so something to be scared about is a fab lab can make a bio lab, a bio lab can make biotechnology, somebody could learn to make a virus. That's scary. That's unlike some of the things I said, I don't worry about, that's something I really worry about, that is scary. Now, how do you deal with that? Prior threats we dealt with command and control. So like, early color copiers had unique codes and you could tell which copier made them. Eventually, you couldn't keep up with that. There was a famous meeting at Asilomar in the early days of recombinant DNA, where that community recognized the dangers of what it was doing and put in place a regime to help manage it. And so that led to the kind of research management. So MIT has an office that supervises research and it works with the national office. That works if you can identify who's doing it and where. It doesn't work in this world we're describing. So anybody could do this anywhere. And so what we found is you can't contain this. It's already out. You can't forbid because there isn't command and control. The most useful thing you can do is provide incentives for transparency. But really the heart of what we do is you could do this by yourself in a basement for nefarious reasons, or you could come into a place in the light where you get help and you get community and you get resources. And there's an incentive to do it in the open, not in the dark. And that might sound naive, but in the sort of places we're working, again, bad people do bad things in these places already, but providing openness and providing transparency is a key part of managing these. And so it transitions from regulating risks as regulation" } ]

[ { "from": "human", "value": "to soft power to manage them. So much potential for good, so much capacity for good that Fab Labs and the ability and the tools of creation" }, { "from": "gpt", "value": "really unlock that potential. Yeah. And I don't say that as sort of dewy-eyed naive. I say that empirically from just years of seeing how this plays out in communities." }, { "from": "human", "value": "I wonder if it's the early days of personal computers though" }, { "from": "gpt", "value": "before we get spam, right? In the end, most fundamentally, literally the mother of all problems is who designed us. So assume success in that we're gonna transition to the machines making machines and all of these new sort of social systems we're describing will help manage them and curate them and democratize them. If we close the gap I just led off with of 10 to the 10 to 10 to the 18 between ChipFab and you, we're ultimately in marrying communication, computation and fabrication, gonna be able to create unimaginable complexity. And how do you design that? And so I'd say the deepest of all questions that I've been working on is goes back to the oldest part of our genome. So in our genome, what are called Hox genes, and these are morphogenes and nowhere in your genome is the number five. It doesn't store the fact that you have five fingers. What it stores is what's called the developmental program. It's a series of steps and the steps have the character of like, grow up a gradient or break symmetry. And at the end of that developmental program, you have five fingers. So you are stored not as a body plan, but as a growth plan. And there's two reasons for that. One reason is just compression. Billions of genes can place trillions of cells. But the much deeper one is evolution doesn't randomly perturb. Almost anything you did randomly in the genome would be fatal or inconsequential, but not interesting. But when you modify things in these developmental programs, you go from like webs for swimming to fingers, or you go from walking to wings for flying. It's a space in which search is interesting. So this is the heart of the success of AI. In part, it was the scaling we talked about a while ago. And in part, it was the representations for which search is effective. AI has found good representations. It hasn't found new ways to search," }, { "from": "human", "value": "but it's found good representations of search. And you're saying that's what biology, that's what evolution has done is creative representation structures," }, { "from": "gpt", "value": "biological structures through which search is effective. And so the developmental programs in the genome beautifully encapsulate the lessons of AI. And this is, it's embodied, it's molecular intelligence. It's AI embodied in our genome. Every bit as profound as the cognition in our brain, but now this is sort of thinking in molecular thinking in how you design. And so I'd say the most fundamental problem we're working on is it's kind of tautological that when you design a phone, you design the phone, you represent the design of the phone. But that actually fails when you get to the sort of complexity that we're talking about. And so there's this profound transition to come. Once I can have self-rebreeding assemblers placing 10 to the 18 parts, you need to not sort of metaphorically, but create life in that you need to learn how to evolve. But evolutionary design has a really misleading trivial meaning. It's not as simple as you randomly mutate things." }, { "from": "human", "value": "It's this much more deep embodiment of AI and morphogenesis. Is there a way for us to continue the kind of evolutionary design that led us to this place from the early days of bacteria, single cell organism" }, { "from": "gpt", "value": "to ribosomes and the 20 amino acids?" }, { "from": "human", "value": "You mean for human augmentation or? For life. What would you call assemblers that are self-replicating and placing parts? What is the dynamic complex things built with digital fabrication? What is that?" }, { "from": "gpt", "value": "That's life. So yeah, so ultimately, absolutely, if you add up everything I'm talking about, it's building up to creating life in non-living materials. And I don't view this as copying life. I view it as deriving life. I didn't start from how does biology work and then I'm gonna copy it. I start from how to solve problems and then it leads me to, in a sense, rediscover biology. So if you go back to Valentina in Ghana making her circuit board, she still needs a chip fab very far away to make the processor in her circuit board. For her to make the processor locally, for all the reasons we described, you actually need the deep things we were just talking about. And so it really does lead you. So let's see. There's a wonderful series of books by Gingery. Book one is how to make a charcoal furnace. And at the end of book seven, you have a machine shop. So it's sort of how you do your own personal industrial revolution. ISRU is what NASA calls in situ resource utilization. And that's how do you go to a planet and create a civilization. ISRU has essentially assumed Gingery. You go through the industrial revolution and you create the inventory of 100,000 resistors. What we're finding is the way the minimum building blocks for a civilization is roughly 20 parts. So what's interesting about the amino acids is they're not interesting. They're hydrophobic or hydrophilic, basic or acidic. They have typical but not extremal properties, but they're good enough you can combine them to make you. So what this is leading towards is technology doesn't need enormous global supply chains. It just needs about 20 properties you can compose to create all technology as the minimum building blocks" }, { "from": "human", "value": "for a technological civilization. So there's going to be 20 basic building blocks" }, { "from": "gpt", "value": "based on which the self-replicating assemblers can work. Right, and I say that not philosophically, just empirically, sort of that's where it's heading. And I like thinking about how you bootstrap a civilization on Mars, that problem. There's a fun video on bonus material for the movie. We're with a neat group of people. We talk about it because it has really profound implications" } ]

[ { "from": "human", "value": "back here on Earth about how we live sustainably. What is that civilization on Mars looks like that's using ISRU, that's using these 20 building blocks" }, { "from": "gpt", "value": "and the self-assembly. Yeah, go through primary, secondary, tertiary, quaternary. You extract properties like conducting, insulating, semiconducting, magnetic, dielectric, flexural. These are the kind of roughly 20 properties. With those, those are enough for us to assemble logic and they're enough for us to assemble actuation. With logic and actuation, we can make micro robots. The micro robots can build bigger robots. The bigger robots can then take the building block materials and make the structural elements that you then do to make construction. And then you boot up through the stages" }, { "from": "human", "value": "of a technological civilization. By the way, where in the span of logic and actuation" }, { "from": "gpt", "value": "did the sensing come in? Oh, I skipped over that. But my favorite sensor is a step response. So if you just make a step and measure the response to the electric field that ranges from user interfaces to positioning to material properties and if you do it at higher frequencies, you get chemistry and you can get all of that just from a step in an electric field. So for example, once you have time resolution in logic, something as simple as two electrodes let you do amazingly capable sensing. So we've been talking about all the work I do. There's a story about how it happens." }, { "from": "human", "value": "Where do ideas come from? That's an interesting story." }, { "from": "gpt", "value": "Where do ideas come from? Where do ideas come from? So I had mentioned Vannevar Bush and he wrote a really influential thing called The Endless Frontier. So science won World War II. The more known story is nuclear bombs. The less well-known story is the Rad Lab. So at MIT, an amazing group of people invented radar which is really credited as winning the war. So after the war, a grand old man from MIT and it was charged with science won the war. How do we maintain that? And the report he wrote led to the National Science Foundation and the modern notion we take for granted but didn't really exist before then of public funding of research or research agencies. In it, he made again what I consider an important mistake which is he described basic research leads to applied research, leads to applications, leads to commercialization, leads to impact. And so we need to invest in that pipeline. The reason I considered a mistake is almost all of the examples we've been talking about in my lab went backwards. That the basic research came from applications and further, almost all of the examples we've been talking about came fundamentally from mistakes. So essentially everything I've ever worked on has failed but in failing something better happened. So the way I like to describe it is ready, aim, fire is you do your homework, you aim carefully at a target you want to accomplish and if everything goes right, you then hit the target and succeed. What I do, you can think of as ready, fire, aim. So you do a lot of work to get ready. Then you close your eyes and you don't really think about where you're aiming but you look very carefully at where you did aim, where you aim after you fire. And the reason that's so important is if you do ready, aim, fire, the best you can hope is hit what you aim at. So let me give you some examples because this is a source of great- You pulled good lines today. Source of great frustration. So I mentioned the early quantum computing. So quantum computing is this power of using quantum mechanics to make computers that for some problems" }, { "from": "human", "value": "are dramatically more powerful than classical computers." }, { "from": "gpt", "value": "You pulled good lines today. Before it started, there was a really interesting group of people who knew a lot about physics and computing that were inventing what became quantum computing before it was clear anything, there was an opportunity there. It was just studying how those relate. Here's how it fits to the ready, fire, aim. I was doing really short-term work in my lab on shoplifting tags. This was really before there was modern RFID. And so how you put tags in objects to sense them, something we just take for granted commercially. And there was a problem of how you can sense multiple objects at the same time. And so I was studying how you can remotely sense materials to make low-cost tags that could let you distinguish multiple objects simultaneously. To do that, you need non-linearity so that the signal is modulated. And so I was looking for material sources of non-linearity and that led me to look at how nuclear spins interact. Just for spin resonance. The sort of things you use when you go in an MRI machine. And so I was studying how to use that. And it turns out that it was a bad idea. You couldn't remotely use it for shoplifting tags, but I realized you could compute. And so with a group of colleagues thinking about early quantum computing, like David DiVincenzo and Charlie Bennett, was articulating what are the properties you need to compute. And then looking at how to make the tags. It turns out the tags were a terrible idea for sensing objects in a supermarket checkout. But I realized they were computing. So with Ike Chuang and a few other people, we realized we could program nuclear spins to compute. And so that's what we use to do Grover's search algorithm. And then it was used for a Shor's factoring algorithm. And it worked out. The systems we did it in nuclear magnetic resonance don't scale beyond a few qubits. But the techniques have lived on. And so all the current quantum computing techniques grew out of the ways we would talk to these spins. But I'm telling this whole story" } ]

[ { "from": "human", "value": "because it came from a bad way to make a shoplifting tag. Starting with an application, mistakes led to the fundamental science. I mean, can you just link on that? I mean, just using nuclear spins to do computation and that like break through. What gave you the guts to try to think through this from a digital fabrication perspective, actually?" }, { "from": "gpt", "value": "How to leap from one to the other? I wouldn't call it guts, I would call it collaboration. So at IBM, there was this amazing group of like, I mentioned Charlie Bennett and David DiVincenzo and Ralph Landau and Nabeel Amir. And these were all gods of thinking about physics and computing. So I yelled at the whole computer industry being based on a fiction, metropolis. Programmers frolicking in the garden while somebody moves levers in the basement. There's a complete parallel history of Maxwell the Boltzmann to Szilard to Landauer to Bennett. Most people won't know most of these names, but this whole parallel history thinking deeply about how computation and physics relate. So I was collaborating with that whole group of people. And then at MIT, I was in this high traffic environment. I wasn't deeply inspired to think about better ways to detect shoplifting tags, but stumbled across companies that needed help with that and was thinking about it. And then I realized those two worlds intersected and we could use the failed approach for the shoplifting tags" }, { "from": "human", "value": "to make early quantum computing algorithms. And this kind of stumbling is fundamental" }, { "from": "gpt", "value": "to the fab lab idea, right? Right, here's one more example. With the student Manu, we talked about ribosomes and I was trying to build a ribosome that worked on fluids so that I could place the little parts we're talking about. And it kept failing because bubbles would come into our system and the bubbles would make the whole thing stop working. And we spent about half a year trying to get rid of the bubbles. Then Manu said, wait a minute, the bubbles are actually better than what we're doing. We should just use the bubbles. And so we invented how to do universal object with little, logic with little bubbles and fluid." }, { "from": "human", "value": "Okay, you have to explain this microfluidic bubble logic. Please, how does this work? It's super interesting." }, { "from": "gpt", "value": "Yeah, and so I'll come back and explain it. But what it led to was we showed fluids could do, it'd been known fluid could do logic, like your old automobile transmissions do logic, but that's macroscopic. It didn't work at little scales. We showed with these bubbles, we could do it at little scales. Then I'm gonna come back and explain it. But what came out of that is Manu then showed you could make a 50 cent microscope using little bubbles. And then the techniques we developed are what we use to transplant genomes to make synthetic life all came out of the failure of trying to make the genome, the ribosome. Now, so the way the bubble logic works is in a little channel, fluid at small scales is fairly viscous. It's sort of like pushing jello, think of it as. If a bubble gets stuck, the fluid has to detour around it. So now imagine a channel that has two wells and one bubble. If the bubble is in one well, the fluid has to go in the other channel. If the fluid is in the other well, it has to go in the first channel. So the position of the bubble can switch, it's a switch. It can switch the fluid between two channels. So now we have one element of switch. And it's also a memory because you can detect whether or not a bubble is stored there. Then if two bubbles meet, if you have two channels crossing, a bubble can go through one way or a bubble can go through the other way, but if two bubbles come together, then they push on each other and one goes one way and one goes the other way. That's a logic operation, that's a logic gate. So we now have a switch, we have a memory, and we have a logic gate, and that's everything you need" }, { "from": "human", "value": "to make a universal computer. I mean, the fact that you did that with bubbles and microfluids, just kind of brilliant." }, { "from": "gpt", "value": "Well, so to stay with that example, what we proposed to do was to make a fluidic ribosome, and the project crashed and burned. It was a disaster. This is what came out of it. And so it was precisely ready-fire aim, in that we had to do a lot of homework to be able to make these microfluidic systems. The fire part was we didn't think too hard about making the ribosome, we just tried to do it. The aim part was we realized the ribosome failed, but something better had happened. And if you look all across research funding, research management, it doesn't anticipate this. So fail-fast is familiar, but fail-fast tends to miss ready and aim. You can't just fail. You have to do your homework before the fail part, and you have to do the aim part after the fail part. And so the whole language of research is about like milestones and deliverables. That works when you're going down a straight line, but it doesn't work for this kind of discovery. And to leap to something you said that's really important is I view part of what the FabLab network is doing" }, { "from": "human", "value": "is giving more people the opportunity to fail. You've said that geometry is really important in biology. What does fabrication biology look like?" }, { "from": "gpt", "value": "Why is geometry important? So molecular biology is dominated by geometry. That's why the protein folding is so important, that the geometry gives the function. And there's this hierarchical construction of as you go through primary secondary tertiary quaternary, the shapes of the molecules make the shape of the molecular machines. And they really are exquisite machines. If you look at how your muscles move, if you were to see a simulation of it, it would look like a improbable science fiction cyborg world of these little walking robots that walk on a discrete lattice. They're really exquisite machines. And then from there, there's this whole hierarchical stack of once you get to the top of that, you then start making organelles that make cells that make organs" } ]

[ { "from": "human", "value": "through the stack of that hierarchy. Just stepping back, does it amaze you that from small building blocks where amino acids, you mentioned molecules, let's go to the very beginning of hydrogen and helium at the start of this universe, that we're able to build up such complex" }, { "from": "gpt", "value": "and beautiful things like our human brain? So studying thermodynamics, which is exactly the question of batteries run out and need recharging, equipment, cars get old and fail, yet life doesn't. And that's why there's a sense in which life seems" }, { "from": "human", "value": "to violate thermodynamics, although of course it doesn't." }, { "from": "gpt", "value": "It seems to resist the march towards entropy somehow. Right, and so Maxwell, who helped give rise to the science of thermodynamics, posited a problem that was so infuriating it led to a series of suicides. There was a series of advisors and advisees, three in a row that all ended up committing suicide that happened to work on this problem. And Maxwell's demon is this simple but infamous problem where right now in this room, we're surrounded by molecules and they run at different velocities. Imagine a container that has a wall and it's got gas on both sides and a little door. And if the door is a molecular sized creature and it could watch the molecules coming, and when a fast molecule is coming, it opens the door. When a slow molecule is coming, it closes the door. After it does that for a while, one side is hot, one is cold. When something is hot and it's cold, you can make an engine. And so you close that, you make an engine and you make energy. So the demon is violating thermodynamics because it's never touching the molecule yet by just opening and closing the door, it can make arbitrary amounts of energy and power a machine. And in thermodynamics, you can't do that. So that's Maxwell's demon. That problem is connected to everything we just spoke about for the last few hours. So Leo Szilard around early 1900s was a deep physicist who then had a lot to do with also post-war anti-nuclear things, but he reduced mass energy and he reduced Maxwell's demon to a single molecule. So the molecule, there's only one molecule and the question is which side of the partition is it on? That led to the idea of one bit of information. So Shannon credited Szilard's analysis of Maxwell's demon for the invention of the bit. For many years, people tried to explain Maxwell's demon by like the energy in the demon looking at the molecule or the energy to open and close the door and nothing ever made sense. Finally, Rolf Landauer, one of the colleagues I mentioned at IBM, finally solved the problem. He showed that you can explain Maxwell's demon by you need the mind of the demon. When the demon opened and closes the door, as long as it remembers what it did, you can run the whole thing backwards. But when the demon forgets, then you can't run it backwards and that's where you get dissipation and that's where you get the violation of thermodynamics. And so the explanation of Maxwell's demon is that it's in the demon's brain. So then Rolf's colleague Charlie at IBM then shocked Rolf by showing you can compute with arbitrarily low energy. So one of the things that's not well covered is that the big computers used for big machine learning, the data centers use tens of megawatts of power. They use as much power as a city. Charlie showed you can actually compute with arbitrarily low amounts of energy by making computers that can go backwards as well as forwards. And what limits the speed of the computer is how fast you want an answer and how certain you want the answer to be. But we're orders of magnitude away from that. So I have a student Cameron working with Lincoln Labs on making superconducting computers that operate near this Landauer limit that are orders of magnitude more efficient. So stepping back to all of that, that whole tour was driven by your question about life and right at the heart of it is Maxwell's demon. Life exists because it can locally violate thermodynamics. It can locally violate thermodynamics because of intelligence and it's molecular intelligence that I would even go out on a limb to say, we can already see we're beginning to come to the end of this current AI phase. So depending on how you count, this is I'd say the fifth AI boom bus cycle. And you can already, it's exploding, but you can already see where it's heading, how it's going to saturate, what happens on the far side. The big thing that's not yet on horizons is embodied AI, molecular intelligence. So to step back to this AI story, there was automation and that was gonna change everything. Then there were expert systems. There was then the first phase of the neural network systems. There've been about five of these. In each case on the slope up, it's gonna change everything. Each case what happens is on the slope down, we sort of move the goalposts and it becomes sort of irrelevant. So a good example is going up, computer chess was gonna change everything. Once computers could play chess, that fundamentally changes the world. Now on the downside, computers play chess, winning at chess is no longer seen as a unique human thing, but people still play chess. This new phase is gonna take a new chunk of things that we thought computers couldn't do. Now computers will be able to do, they have roughly our brain capacity. But we'll keep thinking as well as computers. And as I described, while we've been going through these five boom busts, if you just look at the numbers of ops per second, bits storage, bits of IO, that's the more interesting one. That's been steady and that's what finally caught up to people. But as we've talked about a couple times, there's eight orders of magnitude to go not in the intelligence in the transistors or in the brain but in the embodied intelligence and the intelligence in our body." } ]

[ { "from": "human", "value": "So the intelligent constructions of physical systems that would embody the intelligence versus container within the computation." }, { "from": "gpt", "value": "Right, and there's a brain centrism that assumes our intelligence is centered in our brain. And in endless ways in this conversation, we've been talking about molecular intelligence. Our molecular systems do a deep kind of artificial intelligence, all the things you think of as artificial intelligence does in representing knowledge, storing knowledge, searching over knowledge, adapting to knowledge our molecular systems do. But the output isn't just a thought, it's us. It's the evolution of us. And that's, the real horizon to come is now embodying AI of not just a processor and a robot, but building systems that really can grow and evolve." }, { "from": "human", "value": "So we've been speaking about this boundary between bits and atoms. So let me ask you about one of the big mysteries of consciousness." }, { "from": "gpt", "value": "Do you think it comes from somewhere between that boundary? I won't name names, but if you know who I'm talking about, it's probably clear. I once did a drive, in fact, up to the Mussolini-era villa outside Torino in the early days of what became quantum computing with a famous person who thinks about quantum mechanics and consciousness. And we had the most infuriating conversation that went roughly along the lines of consciousness is weird, quantum mechanics is weird, therefore quantum mechanics explains consciousness." }, { "from": "human", "value": "That was roughly the logical process." }, { "from": "gpt", "value": "And you're not satisfied with that process? No, and I say that very precisely in the following sense. I was a program manager somewhat by accident in a DARPA program on quantum biology. And so biology trivially uses quantum mechanics and that were made out of atoms, but the distinction is in quantum computing, quantum information, you need quantum coherence. And there's a lot of muddled thinking about like collapse of the wave function and claims of quantum computing that garbles just quantum coherence that you can think of it as a wave that has very special properties, but these wave-like properties. And so there's a small set of places where biology uses quantum mechanics in that deeper sense. One is how light is converted to energy in photo systems. It looks like one is olfaction, how your nose is able to tell different smells. Probably one has to do with how birds navigate, how they sense magnetic fields. That involves a coupling between a very weak energy with a magnetic field, coupling into chemical reactions. And there's a beautiful system. Standard in chemistry is magnetic fields like this can influence chemistry, but there are biological circuits that are carefully balanced with two pathways that become unbalanced with magnetic fields. So each of these areas are expensive for biology. It has to consume resources to use quantum mechanics in this way. So again, those are places where we know there's quantum mechanics and biology. In cognition, there's just no evidence. There's no evidence of anything quantum mechanical going on in how cognition works. Consciousness. I'm saying cognition, I'm not saying consciousness, but to get from cognition to consciousness... So McCulloch and Pitts made a model of neurons. That led to perceptrons that then, through a couple of boom busts, led to deep learning. One of the interesting things about that sequence is it diverged off so deep neural networks used in machine learning, diverged from trying to understand how the brain works. What makes them work, what's emerged is, it's a really interesting story. This may be too much of a technical detail, but it has to do with function approximation. We talked about exponentials. A deep network needs an exponentially larger, shallow network to do the same function. And that exponential is what gives the power to deep networks. But what's interesting is the sort of lessons about building these deep architectures and how to train them have really interesting echoes to how brains work. And there's an interesting conversation that's sort of coming back of neuroscientists looking over the shoulder of people training these deep networks, seeing interesting echoes for how the brain works, interesting parallels with it. And so I didn't say consciousness, I just said cognition, but I don't know any experimental evidence that points to anything in neurobiology that says we need quantum mechanics. And I view the question about whether a large language model is conscious as silly in that biology is full of hacks and it works, there's no evidence we have that there's anything deeper going on" }, { "from": "human", "value": "than just this sort of stacking up of hacks in the brain. And somehow consciousness is one of the hacks" }, { "from": "gpt", "value": "or an emergent property of the hacks. Absolutely. And just numerically, I said big computations now have the degrees of freedom of the brain. And they're showing a lot of the phenomenology of what we think is properties of what a brain can do." }, { "from": "human", "value": "And I don't see any reason to invoke anything else. That makes you wonder what kind of beautiful stuff digital fabrication will create. If biology created a few hacks on top of which consciousness and cognition and some of the things we love about human beings was created. It makes you wonder what kind of beauty in the complexity" }, { "from": "gpt", "value": "can create as a digital fabrication. There's an early peak at that, which is there's a misleading term, which is generative design. Generative design is where you don't tell a computer how to design something, you tell the computer what you want it to do. That doesn't work, that only works in limited subdomains. You can't do really complex functionality that way. The one place it's matured though is topology optimization for structure. So let's say you wanted to make a bicycle or a table. You describe the loads on it and it figures out how to design it. And what it makes are beautiful organic looking things. These are things that look like they grew in a forest and they look like they grew in a forest because that's sort of exactly what they are, that they're solving the ways of how you handle loads in the same way biology does. And so you get things that look like trees and shells and all of that. And so that's a peak at this transition to from we design to we teach the machines how to design." } ]

[ { "from": "human", "value": "What can you say about, cause you mentioned cellular automata earlier about from this example you just gave and in general the observation you can make by looking at cellular automata that there's a from simple rules and simple building blocks can emerge arbitrary complexity." }, { "from": "gpt", "value": "Do you understand what that is, how that can be leveraged? So understand what it is, is much easier than it sounds. I complained about Turing's machine making a physics mistake, but Turing never intended it to be a computer architecture. He used it just to prove results about uncomputability. What Turing did on what is computation is exquisite, is gorgeous. He gave us our notion of computational universality and something that sounds deep and turns out to be trivial is it's really easy to show almost everything is computationally universal. So Norm Margulis wrote a beautiful paper with Tom Toffoli showing in a cellular, a cellular automata world is like the game of life where you just move tokens around. They showed that modeling billiard balls on a billiard table with cellular automata is a universal computer. To be universal, you need a persistent state, you need a nonlinear operation to interact them and you need connectivity. So that's what you need to show computational universality. So they showed that a CA modeling billiard balls is a universal computer. Chris Moore went on to show that instead of chaos, Turing showed there are problems in computation that you can't solve, that they're harder than you can't predict. They're actually in a deep reason that they are unsolvable. Chris Moore showed it's very easy to make physical systems that are uncomputable, that what the physics system does, just bouncing balls and surfaces, you can make systems that solve uncomputable problems. And so almost any non-trivial physical system is computationally universal. So the first part of the answer to your question is, this comes back to my comment about how do you bootstrap a civilization? You just don't need much to be computationally universal. So then there isn't today a notion of like fabricational universality or fabricational complexity. The sort of numbers I've been giving you about you eating lunch versus the chip fab, sort of that's in the same spirit of what Shannon did. But once you connect computational universality to kind of fabricational universality," }, { "from": "human", "value": "you then get the ability to grow and adapt and evolve. Because that evolution happens in the physical space" }, { "from": "gpt", "value": "and that's ultimately. Because that evolution happens in the physical space and that's ultimately. And so that's why for me the heart of this whole conversation is morphogenesis. So just to come back to that, what Turing ended his sadly cut short life studying was how genes give rise to form. So how the small amount of it, relatively in effect small amount of information in the genome can give rise to the complexity of who you are. And that's where what resides is this molecular intelligence, which is first how to describe you, but then how to describe you such that you can exist and you can reproduce and you can grow and you can evolve. And so, you know," }, { "from": "human", "value": "that's the seat of our molecular intelligence." }, { "from": "gpt", "value": "The maker revolution in biology. Yeah, it really is. It really is. And that's where you can't separate communication, computation and fabrication, you can't separate computer science and physical science, you can't separate hardware and software." }, { "from": "human", "value": "They all intersect right at that place." }, { "from": "gpt", "value": "Do you think of our universe as just one giant computation? I would even kind of say quantum computing is overhyped in that there's a few things quantum computing is going to be good at. One is breaking cryptosystems, you know how to make new cryptosystems. What it's really good at is modeling other quantum systems. So for studying nanotechnology, it's going to be powerful. But quantum computing is not going to disrupt and change everything. But the reason I say that is this interesting group of strange people who helped invent quantum computing before it was clear anything was there. One of the main reasons they did it wasn't to make a computer that can break a cryptosystem. It was, you could turn this backwards. You could be surprised quantum mechanics can compute or you can go in the opposite direction and say if quantum mechanics can compute, that's the description of nature. So physics is written in terms of partial differential equations. That is an information technology from two centuries ago. The equations of physics are not, this would sound very strange to say, but the equations of physics, Schrodinger's equations and Maxwell's equations and all of them are not fundamental. They're a representation of physics that was accessible to us in the era of having a pencil and a piece of paper. They have a fundamental problem, which is if you make a dot on a piece of paper, in traditional physics theory, there's infinite information in that dot. A point has infinite information. That can't be true because information is a fundamental resource that's connected to energy. And in fact, one of my favorite questions you can ask a cosmologist to trip them up is ask, is information a conserved quantity in the universe? Was all the information created in the big bang or can the universe create information? And I've yet to meet a cosmologist who doesn't stutter and not clearly know how to handle that existential question. But sort of putting that to a side, in physics theory, the way it's taught, information comes late. You're taught about X, a variable, which can contain infinite information, but physically that's unrealistic. And so physics theories have to find ways to cut that off. So instead, there are a number of people who start with a theory of the universe should start with information and computation as the fundamental resources that explain nature. And then you build up from that to something that looks like throwing baseballs down a slope. And so in that sense, the work on physics and computation has many applications that we've been talking about, but more deeply, it's really getting at new ways to think about how the universe works. And there are a number of things that are hard to do in traditional physics that make more sense when you start with information and computation" } ]

[ { "from": "human", "value": "as the root of physical theory. So information and computation being the real fundamental" }, { "from": "gpt", "value": "thing in the universe. Right, that information is a resource. You can't have infinite information in finite space. Information propagates and interacts. And from there, you erect the scaffolding of physics. Now it happens, the words I just said look a lot like quantum field theories, but there's an interesting way where instead of starting with differential equations to get to quantum field theories and quantum field theories, you get to quantization. If you start from computation information, you begin sort of quantized and you build up from there. And so that's the sense in which absolutely, I think about the universe as a computer. The easy way to understand that is just almost anything is computationally universal, but the deep way is it's a real fundamental way" }, { "from": "human", "value": "to understand how the universe works. Let me go a little bit to the personal and the center bits and atoms. You have worked with the students you've worked with have gone on to do some incredible things in this world, including build supercomputers that power Facebook and Twitter and so on. What advice would you give to young people? What advice have you given them? How to have one heck of a great career," }, { "from": "gpt", "value": "one heck of a great life, people. Well, one important one is if you look at junior faculty trying to get tenure at a place like MIT, the ones who try to figure out how to get tenure are miserable and don't get tenure, and the ones who don't try to figure it out are happy and do get it. I mean, you have to love what you're doing and believe in it and nothing else could possibly be what you want to be doing with your life and it gets you out of bed in the morning. And again, it sounds naive, but within the limited domain I'm describing now of getting tenure at MIT, that's a key attribute to it. And in the same sense, if you take the sort of outliers, students we're talking about, 99 out of 100 come to me and say, your work is very fascinating, I'd be interesting to work for you. And one out of 100 come and say, you're wrong. Here's your mistake, here's what you should have been doing. They just sort of say, I'm here and get to work. And again, I don't know how far this resource goes. So I've said, I consider the world's greatest resource, this engine of bright and vene of people of which we only see a tiny little iceberg of it. And everywhere we open these labs they come out of the woodwork. We didn't create all these educational programs, all these other things I'm describing. We tried to partner everywhere with local schools and local companies and kept tripping over dysfunction and find we had to create the environment where people like this can flourish. And so I don't know if this is everyone, if it's 1% of society, what the fraction is, but it's so many orders of magnitude bigger than we see today. We've been racing to keep up with it" }, { "from": "human", "value": "to take advantage of that resource. Something tells me it's a very large fraction" }, { "from": "gpt", "value": "of the population. I mean, the thing that gives me most hope for the future is that population. Once a year, this whole lab network meets and it's my favorite gathering. It's in Bhutan this year because it's every body shape, it's every language, every geography, but it's the same person in all those packages." }, { "from": "human", "value": "It's the same sense of bright and vene of joy and discovery. If there's people listening to this and they're just overwhelmed with how exciting this is, which I think they would be, how can they participate? How can they help? How can they encourage young people or themselves" }, { "from": "gpt", "value": "to build stuff, to create stuff? Yeah, that's a great question. So this is part of a much bigger maker movement that has a lot of embodiments. The part I've been involved in this FabLab network, you can think of as a curated part that works as a network. So you don't benefit in a gym if somebody exercises in another gym, but in the Fab network, you do in a sense benefit when somebody works in another network, another lab in the way it functions as a network. So you can come to cba.mit.edu to see the research we're talking about. There's a Fab Foundation run by Sheri Lassiter at fabfoundation.org. FabLabs.io is a portal into this lab network. fabacademy.org is this distributed hands-on educational program. fab.city is the platform of cities producing what they consume." }, { "from": "human", "value": "Those are all nodes in this network. So you can learn with Fab Academy and you can perhaps launch or help launch" }, { "from": "gpt", "value": "or participate in launching a FabLab lab. Well, in particular, from one to a thousand, we carefully counted labs. Now we're going from a thousand to a million where it ceases to become interesting to count them. And in the thousand to the million, what's interesting about that stage is technologically, you go to a lab not to get access to the machine, but you go to the lab to make the machine. But the other thing interesting in it is we have an interesting collaboration on a FabLab in a box. And this came out of a collaboration with SolidWorks on how you can put a FabLab in a box, which is not just the tools, but the knowledge. So you open the box and the box contains the knowledge of how to use it as well as the tools within it so that the knowledge can propagate. And so we have an interesting group of people working on, the original FabLabs, we have a whole team to get involved in the setting up and training. And the Fab Academy is a real in-depth deep technical program in the training. But in this next phase, how sort of the lab itself knows how to do the lab. We've talked deeply about the intelligence in fabrication, but in a much more accessible one about how the AI in the lab in a fact becomes a collaborator with you in this nearer term to help get started. And for people wanting to connect, it can seem like a big step, a big threshold, but we've gotten to thousands of these and they're doubling exactly that way" } ]

[ { "from": "human", "value": "just from people opting in. And in so doing driving towards this kind of idea of personal digital fabrication." }, { "from": "gpt", "value": "Yeah, and it's not utopia, it's not free, but come back to today we separately have education, we have big business, we have startups, we have entertainment, sort of each of these things are segregated. When you have global connection to one of these local facilities, in that you can do play and art and education and create infrastructure. You can make many of the things you consume. You could make it for yourself, it could be done on a community school, it could be done on a regional scale. I'd say the research we spent the last few hours talking about I thought was hard. And in a sense, I mean, it's non-trivial, but in a sense, it's just sort of playing out we're turning the crank. What I didn't think was hard is, if anybody can make almost anything anywhere, how do you live, how do you learn, how do you work, how you play? These very basic assumptions about how society functions. There's a way in which it's kind of back to the future in that this mode where work is money is consumption and consumption is shopping by selecting is only a kind of a few decade old stretch. In some ways, we're getting back to a Sami village in North Norway is deeply sustainable, but rather than just reverting to living the way we did a few thousand years ago, being connected globally, having the benefits of modern society, but connecting it back to older notions of sustainability. I hadn't remotely anticipated just how fundamentally that challenges how a society functions and how interesting and how hard it is to figure out" }, { "from": "human", "value": "how we can make that work. And it's possible that this kind of process" }, { "from": "gpt", "value": "will give a deeper sense of meaning to each person. Let me violently agree in two ways. One way is this community making crosses many sensitive sectarian boundaries in many parts of the world where there's just implicit or explicit conflict, but sort of this act of making seems to transcend a lot of historical divisions. I don't say that philosophically, I just say that as an observation. And I think there's something really fundamental in what you said, which is deep in our brain is shaping our environment. A lot of what's strange about our society is the way that we can't do that. The act of shaping our environment touches something really, really deep that gets to the essence of who we are. That's again why I say that in a way the most important thing made in these labs is making itself." }, { "from": "human", "value": "What do you think if the shaping of our environment gets to something deep, what do you think is the meaning of it all?" }, { "from": "gpt", "value": "What's the meaning of life, Neel? I can tell you my insights into how life works. I can tell you my insights and how to make life meaningful and fulfilling and sustainable. I have no idea what the meaning of life is," }, { "from": "human", "value": "but maybe that's the meaning of life. The uncertainty, the confusion, because there's a magic to it all, everything you've talked about from starting from the basic elements with the big bang that somehow created the sun, that somehow set a few to thermodynamics and created life and all the ways that you've talked about from ribosomes that created the machinery, that created the machine and then now the biological machine creating through digital fabrication, more complex artificial machines. All of that, there's a magic to that creative process. We humans are smart enough to notice the magic." }, { "from": "gpt", "value": "So it's, you haven't said the S word yet. Which one is that? Singularity. Yeah, I'm not sure if Ray Kurzweil is listening, if he is, hi, Ray, but I have a complex relationship with Ray because a lot of the things he projects, I find annoying, but then he does his homework and then somewhat annoyingly, he points out how almost everything I'm doing fits on his roadmaps. Yeah. And so the question is, are we heading towards a singularity? So I'd have to say I lean towards sigmoids rather than exponentials. We've done pretty well with sigmoids. Yeah, so sigmoids are things grow and they taper and then there can be one after it and one after it. So, I'll pass on whether there's enough of them that they diverge. But the selfish gene answer to the meaning of life is the meaning of life is the propagation of life. And so, it was a step for atoms to assemble into a molecule, for molecules to assemble into a protocell, for the protocell to form. To then form organelles, for the cells to form organs. The organs to form an organism. Then it was a step for organisms to form family units. Then family units to form villages. You can view each of those as a stack in the level of organizations. So, you could view everything we've spoken about as the imperative of life, just the next step in the hierarchy of that and the fulfillment of the inexorable drive of the violation of thermodynamics. So, you could view, I'm an embodiment of the will" }, { "from": "human", "value": "of the violation of thermodynamics speaking. The two of us having an old chat, yes. Yeah. And so continues and even then the singularity" }, { "from": "gpt", "value": "is just a transition up the ladder, yes. There's nothing deeper to consciousness than it's a derived property of distributed problem solving. There's nothing deeper to life than embodied AI in morphogenesis. So, why so much of this conversation in my life is involved in these fab labs? And initially it just started as outreach, then it started as keeping up with it. Then it turned to, it was rewarding. Then it turned to we're learning as much from these labs in as goes out to them. It began as outreach, but now more knowledge is coming back from the labs than is going into them. And then finally, it ends with what I described as competing with myself at MIT, but a better way to say that is tapping the brainpower of the planet. And so, I guess for me personally, that's the meaning of my life." } ]

[ { "from": "human", "value": "And maybe that's the meaning for the universe too. It's using us humans and our creations to understand itself in the way it's, whatever the creative process" }, { "from": "gpt", "value": "that created Earth, it's competing with itself. Yeah, so you could take morphogenesis as a summary of this whole conversation, or you could take recursion that in a sense," }, { "from": "human", "value": "what we've been talking about is recursion all the way down. And in the end, I think this whole thing is pretty fun. It's short, life is, but it's pretty fun. And so is this conversation. I mentioned to you offline, I'm going through some difficult stuff personally, and your passion for what you do is just really inspiring. And it just lights up my mood and lights up my heart. And you're an inspiration for, I know, thousands of people that work with you at MIT and millions of people across the world. It's a big honor that you sit with me today." }, { "from": "gpt", "value": "This was really fun." }, { "from": "human", "value": "This was a pleasure. Thanks for listening to this conversation with Neil Gershenfeld. To support this podcast, please check out our sponsors in the description. And now, let me leave you with some words from Pablo Picasso. Every child is an artist. A challenge is staying an artist when you grow up." }, { "from": "gpt", "value": "Thank you for listening and hope to see you next time." } ]

[ { "from": "human", "value": "As always, no ads in the middle. I try to make this interesting, but if you must skip them, please still check out our sponsors. I enjoy their stuff. Maybe you will too. This episode is brought to you by Asleep and its new Pod3 mattress. The very device, the technological smart portal into a dimension that I take every single day for many hours once and for a few minutes once or twice a day, also known as NAPs. Wherever the place that that portal takes me, this is a wonderful place. All I know, very importantly, is when I return, I return a new human being." }, { "from": "gpt", "value": "Whatever happens to me in the middle," }, { "from": "human", "value": "all I know are the creatures that dwell in that alternate dimension. Whatever they do to me, they reset the CPU and the memory and whatever is required. They defrag the hard drive enough to where I wake up refreshed and ready to take on the rest of the day. NAPs are just magic. Big believer, big lover of NAPs. If there's a hack to the simulation that we're all existing in, NAPs are it. Anyway, check out this magical portal into the alternate dimension and get special savings when you go to eighthsleep.com slash Lex. This episode is also brought to you by Linode, now called Akamai, and they're incredible Linux virtual machines. I've said this many times, but Linux is my favorite operating system, all of the flavors. Some people talk about their favorite flavors of ice cream. I talk about my favorite flavors of Linux, both the ice cream and the Linux that have evolved over the years. There was a time when I used to love coconut ice cream, and there was a time when I would use Gentoo, but that time is long gone. I've grown up a little bit, still young in heart, but practical enough to be using a Bantu. And my favorite flavor of Bantu is mate. Anyway, all of that doesn't matter. The point is Linux is Linux, and you can use Linux in the cloud in a way that's super easy for small personal projects and for huge systems, great customer support, lower cost than AWS. I love Linode, now called Akamai. I need to get used to saying I love Akamai. I rose by any other name will smell as sweet. Shakespeare has some wisdom on that one. Anyway, check out this beautiful, sweet-smelling rose that is Akamai at akamai.com slash Lex, or linode.com slash Lex for a free credit. This show is also brought to you by Insight Tracker, a service I use to track data that comes from my own body, and that data, when run through machine learning algorithms, is able to give me advice on what to do with my life. Well, some basic sort of lifestyle gone, and diet decisions. Unfortunately, you can't tell me what I should do in terms of career or relationships or how to navigate all the different and the beautiful complexities of life. I think using data that comes from your own body to make decisions about your life, health decisions, medical decisions, career decisions, all that kind of stuff, obviously when you look at the power of large language models, it's clear that when paired with great data, large neural networks can help do incredible predictive work. This is the future, but it starts with getting data from your body and starts by taking action based on that data. And so Insight Tracker is a pioneer that I support in this effort that I think will define a great quality of life in the 21st century. So you should check them out and get special savings at a limited time when you go to insighttracker.com slash Lex. This is a Lex Friedman podcast to support it. Please check out our sponsors in the description. And now, dear friends, here's Randall Kennedy. You wrote a book whose title is the N word spelled out with a hard R at the end. So let's start with the history of this word." }, { "from": "gpt", "value": "What is the history of this tree of the N word? The word you're referring to is nigger. The book that you're referring to is nigger, the strange career of a troublesome word. The word dates back to the 16th, 17th century. It's got a long lineage in other words. Basically Latin, basically Spanish, basically NIG, you know, black in various formulations. We don't know actually how the term nigger became a slur. So there were words that were close to NIGGER that were used in various ways. For instance, NIGG, UH has been, was used. NIGG, U-R is used. And sometimes it was used in a way that seemed to be just purely descriptive. We do know that by the early 19th century, it had become a slur. It had become a derogatory word about which people complained." }, { "from": "human", "value": "But exactly how that came about, not altogether clear. So it's been 20 years since you've written the book. What have you, what wisdom have you gained about this word since writing the book? And maybe having to interact with people, having to read, having to see," }, { "from": "gpt", "value": "having to feel the response to the book. This book has generated a lot of controversy. I thought it would, it's probably generated more controversy than I had anticipated. It has certainly generated more, more different sorts of experiences than I had anticipated. So for instance, I did not think that writing this book would prompt people to ask me to be an expert witness in cases. And over the past 20 years, I've been an expert witness in a number of different cases. I've been an expert witness in case, in a murder case, in various cases of assault. I've been an expert witness in cases involving tort cases, intentional infliction of emotional distress. I've been an expert witness in a number of employment cases. I had not anticipated that, nor had I anticipated the extent to which people would get in trouble for using my book. Every year, there are teachers who are suspended or who are fired because they will excerpt a chapter of my book. Let's imagine, and this is not, I'm not imagining things as has happened. A teacher is teaching, for instance, the Mark Twain's Huckleberry Finn. The word nigger appears in that book over 200 times. The teacher, trying to be earnest, trying to be sensible, trying to be serious, will excerpt a part of my book to acquaint students with the history of the word and maybe the history of controversy involving the use of the word in this particular novel. The teacher will give it out, hand it out to the teachers, hand it out to students, and there have been a number of teachers who've been suspended or worse because of that. Students will get upset, go home, tell their parents. Their parents will storm to the school and say that this is terrible. The teacher is, quote, using the word nigger in an offensive way, and oftentimes, administrators will basically abandon the teacher. And whenever this comes to my attention, I write, I'll write the superintendent of schools or I'll write the principal, or sometimes I'll write an opinion editorial piece for the local newspaper. But every year, there are teachers who are disciplined for using my book." } ]

[ { "from": "human", "value": "I had not anticipated that. And what is the nature of the letter or the op-ed that you write on why they shouldn't be disciplined" }, { "from": "gpt", "value": "or if they, to the degree they should be or shouldn't be? There's not been one case, there's not been one case that has come to my attention in which it was even remotely sensible for the teacher to be disciplined. And what I say is that, number one, frankly, I go through, what I write is almost a synopsis of my book. Number one, this is an important word in American history. It is a word that is explosive, that's why people get so upset. It's a word that's volatile, it's a word that has typically been used in a terrible way. It's a word that is part of the soundtrack of racial terrorism in the United States. So people ought to know about this word. I mean, if you're interested in knowing the real history of the United States, if you're interested in knowing about lynching, if you're interested in knowing about the way in which black people have been terrorized in the United States, you need to know this word. You need to know that history. So you need to know why it is that people are upset about the word. But it doesn't end there. You have to know that. And if you know that, then that knowledge should equip you to be careful. It should equip you to know that, to know the range of context in which this word appears. But again, it doesn't just end there because especially young people, you tell that to young people, they nod, they read, they understand that. But then what? But then they turn on their radios and they turn on, they listen to Spotify. They listen to some of their favorite entertainers. They listen to Dr. Dre. They listen to the ghetto boys. They listen to Snoop. And they listen to NWA. What are they here? They listen to standup comedians. They listen to Dave Chappelle. They listen to Cat Williams. What do they hear? They hear the word nigger or nigga being used in a lot of different ways. And so they need to know about that as well. What are people doing? How does one explain the fact that Dick Gregory, Dick Gregory was a comedian, activist, friend of Martin Luther King Jr., a true activist? I mean, he had a flourishing career as an entertainer that he abandons in order to struggle for racial justice throughout the United States and including the Deep South. How does one explain the fact that he wrote several memoirs but his first memoir is called Nigger, a memoir. How does one explain that? How does one explain the way in which, how does one explain Richard Pryor? I mean Richard Pryor's best album is that niggers crazy. Well, was Richard Pryor trying to put down Black people? How does one explain that? One can only explain that by getting deeper into the word, by understanding that, yes, this is a word that has been used in a derogatory way. This is a word that has been used to put people down. This is a word that has been used to terrorize people. You gotta know that, but you also have to know that this is a word that has also been put to other uses. There are artists, there are entertainers who have used this word, like Dick Gregory used it, to put up a mirror to American society and say, look at this word and look at the terrible way in which it's been used. We don't want you to look away. No, don't look away. No euphemism, no asterisk, no n-word, no nigger. Now we want you to look at that and we wanna talk about that. James Baldwin, there was a documentary about James Baldwin a couple of years ago, highly lauded documentary. The title that was given in this documentary was, I Am Not Your Negro. That's not what James Baldwin said. Anybody can go to YouTube right now, take a look. James Baldwin said, I am not your nigger. And then he went on to talk about that. Well, James Baldwin wasn't, again, he wasn't starting to cover up anything. He wanted people to face the facts of American life. And it seems to me that if you're a teacher and you want to have your students face the facts about American life, well, you've got to grapple with the word nigger. Now, let me just quickly say, teachers have a tough job. And if we're talking about students, of course, there's a wide range of students. Am I saying that one ought to give my book to kindergartners? No, kindergartners are probably not ready for such a book. Third graders, probably no, not third grade. If we're talking however, about people in the 10th grade, do I think the 10th graders can read my book? Yeah, sure, absolutely. 11th graders, 12th graders, people in college, there are people in college, there are people in college, there are people in law schools. Teachers in law schools have been disciplined for, because the word nigger has come out of a teacher's mouth. Why? A couple of cases recently, a teacher would be reading a court opinion. The word appears in the court opinion, the teacher pronounces the word, students get up, leave in a huff, report the teacher. There's some instances in which teachers have been under those circumstances, have been disciplined. In my view, that's bad. And people ought to say it's bad. It's bad pedagogically. And frankly, in many of these instances, it's not only bad, it's stupid. And I don't mind saying that. I think that some of these instances in which teachers have been disciplined, absolutely stupid. People say, well, the teacher used the word, excuse me, used the word. It'd be one thing, it'd be one thing if a teacher looked at a student and called the student nigger. Get out of here, nigger. That'd be one, that'd be, you know, fine. Discipline, that's teacher, that's bad, but that's not what's going on. You don't have, none of these are cases in which you have an individual who is a stranger to another individual and this word just sort of comes out, no. What we have here is a class involving a person who is a teacher interacting with students, talking about subjects in which it would be perfectly understandable why this word would emerge as a subject of conversation. Now, under those circumstances, it's somehow wrong for a teacher to utter this word. In my view, the answer is no. And, you know, I said that 20 years ago," } ]

[ { "from": "human", "value": "I say it even more emphatically now. Still, it is one of the most powerful words in the English language. And there's a kind of responsibility that we as humans should have with words, with statements. That word, if not used skillfully, if not used competently, even when just read from a legal transcript," }, { "from": "gpt", "value": "can do more harm than good. Yeah, I agree with what you say. Yes, words are powerful. Words do matter. And so I am certainly not suggesting that people be lax. I'm not suggesting that people be irresponsible. It's precisely because words matter, however, that we need to be willing to face words and grapple with words and talk about words and talk about the history of words, precisely because words matter. And among other things, it seems to me, it's important to understand that words can mean different things in different contexts. It's not the case that a word means the same thing in every context, the word discriminating. Sometimes it's a very bad thing. That person discriminates. And when, again, intonation of voice means something. If I say that person discriminates, and I'm obviously being disapproving, implicitly what I'm saying is that person distinguishes between things on an unjustifiable basis. And that's a bad thing. On the other hand, that person has discriminating taste. Oh, that means something very different. That means that the person differentiates in a way that shows that they understand the difference between excellent, good, and not so good. And we think that that's a good thing. So words can mean different things in different contexts. It seems to me that that's something that actually we ought to recognize and talk about." }, { "from": "human", "value": "Well, some words enter this territory of being a slur. And it seems like when they cross the line into being a slur, there's the number of contexts in which it's okay to mention it exponentially decreases. Right?" }, { "from": "gpt", "value": "Uh, no, no, I'm gonna resist that a little bit because the whole idea of slurs, slurs change. Yankee was a slur. Yankee was a slur in 18th century United States. Slur today, New York Yankees, I'm a Yankee fan. I'm a Yankee, Queer, queer, and this is in my lifetime. There was a time you queer and people would really run away from it and that was a bad thing. And then, thank goodness, I'm a queer fan. Goodness, gay liberation movement, gay liberation movement, basically, we're not gonna run away from this. We're gonna grab this quote slur and we're going to affix it to ourselves. And we are going to repurpose it. Now, the word queer is again, you know, can it be a slur? Yeah, it can be a slur, doesn't have to be. And it seems to me that it's important for people to know about how a word, a symbol, in some context can be a slur, in some context, doesn't have to be. So the whole idea of what's a slur," }, { "from": "human", "value": "that's a complicated idea in and of itself. It's very complicated. It's, if I may say, almost fascinating how language evolves. But if we were to kind of have a minute by minute evaluation of the most powerful, intensely slur-like words in the world, I think the N word with a hard R at the end, which is the title of your book," }, { "from": "gpt", "value": "is number one on that list. Well, I probably so. And of course, that's one of the reasons" }, { "from": "human", "value": "why I wrote a little book about it. Yeah, but it hasn't, even since you wrote a little book about it, it seems like it's maintained its number one status. You mentioned queer. Mm-hmm. Maybe queer was in the top 20, I don't know, for a while, and now it's sliding into the top 1,000," }, { "from": "gpt", "value": "and the N word is at the top. You're absolutely right. The origins of this book, I clearly remember, I was at my office and I was thinking about lecture topics. And I get invited to give lectures from time to time. And I was thinking, well, what might make for an interesting lecture? And all of a sudden, the word nigger popped into my mind. Now, this is a word, I've grown up with this word. I mean, there's never been a time in my life when, at least in my conscious life, in which this word's been absent. I mean, in my household, for instance, in my household, my parents are black people, my parents were refugees from the Jim Crow South. I was born in the Deep South, South Carolina. In my household, I heard the word nigger used in every possible way. I heard it used as a slur. I also heard it used with respect to people who were praised, you know? My father, I clearly remember my father, whom I revere. That's the smartest nigga in the world. That's the bravest nigga in the world. That's the baddest nigga I know. He wasn't putting people down. This is the way he talked. And I grew up hearing this word in various ways. And so I was thinking to myself, God, where did this word come from? And one of the first things I did, I clearly remember just jumping up out of my seat, running up to the library, Oxford English Dictionary. When did this word first appear in English? Was the history of the word. And then what really sort of grabbed my attention is I went, I get my computer going, and I asked the computer system, give me every case, every federal court case in which this word appears, thousands of cases. And then I said, oh my goodness, this is really, and you know, this is really interesting. And then I started just cataloging all the different cases. There came a point, I'd say probably about a month into this, I compared the usage, I compared the number of times a nigger came up with other sorts of slurs. So for instance, kike, K-I-K-E, long time, derogatory word for Jews. How many times has this word come up? There was a time in which the word appeared, but nothing like the infamous N word, nothing. What about wetback? What about, and then I just, let me just take a look at all the other slurs. Nothing came close, not even remotely close to nigger. And I think it has something to do with, I think it has something to do with the uniqueness of the color line, particularly as it pertains to African Americans. I think that the fact that nigger sort of occupies such a unique status among slurs, I think that's a reflection of the unique stigma" } ]

[ { "from": "human", "value": "that has been imposed on African Americans. It's hard to know the chicken or the egg, why one word is able to so distinctly and clearly encapsulate this struggle between races that is throughout American history. I mean, they didn't have to probably be so," }, { "from": "gpt", "value": "but it came to be that way. It became that, not only that, not only that, but of course the nigger spurred other slurs. So Arabs, sand niggers, the Irish, the niggers of Europe," }, { "from": "human", "value": "women, the niggers of the world." }, { "from": "gpt", "value": "John Lennon even has a song. That's right. Well, I think Yoko Ono I think had something to do with that song. So, I mean, it is a slur that has spawned other slurs. And again, that's why, as you indicated a moment ago," }, { "from": "human", "value": "this is a quite unique term. But are you conscious? Are you deliberate in you saying this word? So let me just say from a personal experience, maybe my upbringing, where I came from, in my daily life, I don't think I've ever heard that word with a hard R said as often used clearly in my life, said that I've heard it today more than I have ever heard in my entire life. And I think there's a few people who listen to this, they'll be listening to this and be very uncomfortable. I would say not in a bad way, probably in a good way. I'm uncomfortable now, and I am almost introspecting and trying to figure out why I'm uncomfortable. And I think even the title of your book is making me think that, just looking into my own mind and trying to understand, wow, words of power. And why does it have so much power? But are you deliberate in that action? And by the way, not only are the people listening to this sweating, this will be on YouTube in part. And YouTube, the people on the other side" }, { "from": "gpt", "value": "will be sweating, what do we do with this? Yeah, well, am I deliberate? The answer is yes. And let me unpack that a little bit. First, that's right, I mean, am I deliberate? Yeah, I deliberately wrote a book called, Nigger, The Strange Career of a Troublesome Word. And was that deliberate?" }, { "from": "human", "value": "Yeah, that was quite deliberate." }, { "from": "gpt", "value": "But the title could have been N word, versus spell. The title could have been N word, sure. The title could have been... The title could have been a book about a word that causes pain to many people. I could have named it that. There are many titles, I could have used. Did I want a title that would be provocative? Did I want a title that would grab people? The answer to that is, yes. I'm a writer, I want people to read what I write. Was I being sensationalistic? Well, I mean, if you wanna put it like that, yes. I don't, I'm not embarrassed to say that. I mean, I'm sure that when people write books, they think really hard about their titles and they try to get a title that will grab people's attention. I know people, respect people very deeply who never, it's a matter of principle, never utter this word. And I've talked with people, I've had people say, I read your book, let's talk about it, but let's be very clear, I'm not gonna use the word. I've had many conversations with people who've asked me not to use the word. I was on a, the first time this came up was a book tour. It was 20 years ago when the book first came out and I was on one of these call-in shows early in the morning. Time came to call in, seven o'clock, I call in at five or seven. Right before I go on, the host of the show says, oh, by the way, we have a strict policy here at the station. We never use this word. And I said, well, gosh, I wish you had told me this earlier. Does this mean then that you're never going to pronounce fully the title of my book? And she said, that's right. And I had to make up choice right then and there. Am I gonna go on or am I not? What'd you do? I went on and I abided by the station's rules and you know, fine, we had a perfectly fine conversation. And you know, there is a place for euphemism. The American language is a very supple language. There are lots of words that one can use. I do not get angry with people who don't, you know, they say as a matter of principle, they're not gonna use the word fine. I'm willing to, I understand where they're coming from" }, { "from": "human", "value": "and often I will defer to their wishes." }, { "from": "gpt", "value": "What'd you do? All I say is I want people to understand where I'm coming from. I'm not just using this word willy-nilly. There's a pedagogical reason. There is a reason for why I'm saying what I'm saying." }, { "from": "human", "value": "There is a reason why I use the word. Can you make the case why using the word is a good idea? And can you make, can you steal man the case why it's a bad idea? Maybe you've heard from some critics. Yes. Who said that you saying this word out loud is actually causing a lot of harm. Not like harm because people's feelings are hurt but increasing the amount of racism and hate in the world." }, { "from": "gpt", "value": "Yeah. Critics. Let's start with the critics. One, again, going back to when the book was first published, I remember going to the first bookstore I went to and I talked about the book, had a very, you know, talked, asked questions. And the last comment wasn't a question but the last comment was made by an elderly black man. I called on him and he said, I've listened to what you've had to say and I appreciate what you've had to say. But he, he said, but I remain unconvinced and I remain unconvinced because when I was coming up, this word was used to put me in the back of the bus. And this word was used to prevent me from voting. And this was the word that was used to justify me never being called as a juror. So to me, this word has only one meaning. It's a terrible meaning. I'm never going to use the word. It hurts me when I hear people use the word, especially those who don't know anything about the, really about the history of it. And he went on to say, I think that your book though well-intended he's probably going to be seen by some people as giving them permission to use the word. And then he stopped. And I thought there was a lot of power behind that gentleman's comment. I think that what he said is probably correct. And so far as there are probably some people who read the book, there probably some people who are listening to our conversation right now who will think that I'm giving people permission to use the word. You know, I have said that I'm not, I want people to understand the word, I think that there is a burden that comes from whenever you utter a word like this, but that's, you know, that that's a critique and I think there's strength to that critique. I'm not gonna say that that's a ridiculous critique. I think that there is something to that. And by the way, I should say, and that's why I would say to anyone, that's right, if this word comes out of your mouth, you are taking on real responsibility. So for years, it doesn't happen so much now, but they were, I'd say for about the first five years after this book was published, I would get an email at least once a week and it would begin like this, you know, dear professor Kennedy, I read your book and I'm calling to ask you a question. And as soon as I saw that, I knew what the question was gonna be and then what the person would say is the following, I like rap. And then I knew it was coming, I like rap. I'm white and I have black friends and we listened to rap and we're driving in the car and we're listening to the song, we start humming along and singing along and my black friends sing along and when nigger or nigga comes up, they sing and I don't know what to do. Is it wrong for me to sing along? This happened so often that I'd say about the 10th time I got in such an email, I wrote a form letter because I didn't wanna just, you know, take up time writing, you know, sort of crafting letter after letter after letter. So I wrote a form letter and basically what I said was listen, number one, I'm flattered that you're asking me but number one, you should have a conversation with your friend, number two, no matter what your friend says let me put something else for you to consider. Let's suppose for the sake of discussion that your friend says, oh, doesn't bother me, I know where you're coming from, we're just enjoying the music, I don't think that this is a you know, a racist utterance, you know, coming out of your mouth. Let's suppose that your friend says that, that doesn't end matters because let's imagine the following, let's imagine that you're in a theater and you're waiting for the, you know, a film to start and you're just, you know, talking with your friends are singing with your friends or just, you know, kicking back with your friends and they're talking about nigger this or, nigger this and you say it, you the white boys say it and the next thing you feel is a fist, a big fist in your mouth that has been launched by a person that you did not see who was right behind you. All this person saw was a white person saying nigger and the next thing, that's not, you know, some sort of overheated scenario coming from some law professor's mind. That is a very plausible scenario. So you have to be worried about lots of things, including mistake. So my advice to you is be prudent. I would stay clear of the word unless, unless you're very certain and unless if you're called on it, you feel you're in a position to defend yourself, defend what you're doing." } ]

[ { "from": "human", "value": "But the prudent thing would be to stay clear. There's so many questions I wanna ask there. One is about the violence and the legal aspect of that. It's very interesting, you raised that in the book. But, you know, I do wanna bring up something I probably disagree with you on, which is you say that there's not a significant difference between the different variations of the n-word. The one the, or maybe you don't. I just listened to a bunch of your interviews. So there's the version with the er at the end, version with the ga at the end and then gro at the end. These are all different versions. And I feel like in that list of powerful words, you know, I feel like there's a distinction. I feel that the number one spot is the one with the hard R. And I don't know, maybe you can try to shed light, but I feel like the one that ends in ga is really far down the list in terms of modern culture. So this is, we talked about the evolution of the words and the word queer, for example. It feels like because maybe because of rap, because of comedians, because it's become much more, it lost so much of its power. Oh, you don't think so?" }, { "from": "gpt", "value": "No, I think there's a difference between nigger and nigger. I mean, people make a distinction between them. And I think that, to the extent that lots of people make a distinction between them. I think, you know, just as a sociological fact, they are different. I think that people who get upset if somebody, especially white people, so, you know, if a white person says nigga, and they're, you know, and they're, you know, sort of criticized about it, and they say, well, I didn't say nigger, I said nigga, believe me. I think most people who are mad at them are gonna stay mad at them. Now you raised the word, you know, so nigger and nigga, I would put in a very different category than nigro. Educate me here. Well, yeah, sure, I'm happy to. Nigro is a, also controversial. It's also controversial. But nigro has never been viewed by a substantial number of people as a derogatory term, at least in the, with the same amount of animus, the same amount of, it's a very different kind of word than nigger or nigga. I mean, after all, I mean, you know, nigro, Martin Luther King, Jr., you know, all of his great oratory, nigro. You read the work of the great W.E.B. Du Bois, nigro. You read the work of, my boss, so for instance, I use the word nigro. I use African American, black, Afro American, but I also use the word nigro. Now there are some people who get really mad at me because of, you know, when I use the word nigro. And then, so for instance, there's students who've gotten really quite exercised and they'll say, you know, I'll be giving a lecture and a hand will go up, I'll call on somebody and I'll say, listen, are you using the word nigro in its purely because of the historical time period that you're using? So, you know, is that why you're using it or are you using it in your own voice? And often I'll say, well, I'm using it in my own voice. And I'll say, well, I'm offended. We think that this is, you know, that's old timey. It's derogatory. When this first came up, I said, let's pause for a moment and I'll take that under advisement and let me look into this. And I ended up writing an essay about it, an essay about the history of the terms that black people have used to describe themselves. And it's a long list, you know, black, colored, Afro-American, African-American, nigro, et cetera. So I go through all that and I said, now, let me just tell you, I know for certain when I started using the word nigro often in writing, I can date it, 1983, the summer of 1983 is when I started using the word nigro in my professional life as a lawyer. And I did it for a very specific reason. I did it because my boss demanded that I nigro capital N. Now, who was my boss? My boss in the summer of 1983 was Associate Justice of the United States Supreme Court Thurgood Marshall, known as Mr. Civil Rights. Now it seems to me you were telling me that this word is so out of bounds, that this word is derogatory, nobody should ever use this word. Does the fact that Thurgood Marshall demanded that I use this word, does that complicate things a little bit? And so I think that people, again, you know, ought to know more. I mean, I've encountered students who don't know very much but who want to lecture me on word usage because they know, you know, three sentences about current fashion and hold it, you know, hold it. By the way, I push it further. I sometimes use the word colored. And then, you know, some people really don't like that, really don't like that. Colored, well, you know, there's an organization still very much alive in American life and law. The National Association for the Advancement of Colored People, the N-A-A-C-P. They haven't changed their name. And as far as I'm concerned, it's a wonderful organization. There are people who have used the word colored. That's what my grandmother used. Colored, perfectly fine as far as I'm concerned. So again, there are lots of different words that one can use. You can use different formulations. Understand that people have different preferences, fine." } ]

[ { "from": "human", "value": "I have my preferences, at least know where I'm coming from. Still, words have power and they have power to hurt. And there's a lot of reasons that I could see to justify the use of the word in its full form, as you're saying, in this conversation as you're using it. One of them is perhaps fighting for the freedom to be able to use those kinds of words. So let me ask you about the freedom of speech and the censorship of the word. Should the use of the N-word be censored, for example, on social networks? So we can come up with different places. We can say university campuses, maybe in op-eds or, I don't know, but I think social networks currently is a very interesting place. There's a lot of conversations that are happening on them. There's the ability, the technical capability to sentence, to remove the ability of people to use the word. Do you think they should be allowed to use the word" }, { "from": "gpt", "value": "on Twitter, for example? My response is it all depends on the way in which the word is being used. If the word is being used to intimidate, if the word is being used to terrorize, then no, along with lots of other words, by the way. I could use the word, I suppose, gentlemen. I mean, I'm sure that I could conjure up a way in which that word could be used in an intimidating way. So would I be happy if there was a technology that always blanked this word, N-I-G-G-E-R? I would be against that. For one thing, it would erase the name of one of my books. I think that the book actually has a lot of useful information, makes some useful points. Let's imagine a world in which there was a technology that blanked every time N-I-G-G-E-R appeared. You would have, what would that do to novels by James Baldwin, by Toni Morrison, by, what would that do to the speeches of Malcolm X, Martin Luther King, Jr.? What would that do to the comedy albums of Richard Pryor, Cat Williams, Dave Chappelle? If one sort of plays that out, would one want to have all of those blanks in important literary and political performances? No, I don't want such blanks. What I want, I would want the word to be there and for people to understand how to deal with this word. And by the way, you've used the word hurt an awful lot. So let's talk about hurt. I think we need to be more careful with the way in which we deal with hurt. Because people can be justifiably hurt, you can have justifiably hurt feelings. And if somebody has justifiably hurt feelings, I then think that we should turn to the person who has hurt those feelings and say, you have acted wrongly because this person's feelings are justifiably hurt in relationship to what you've done. On the other hand, there are people who have hurt feelings and frankly, it's unjustified. So just imagine the following. Let's imagine that I give a talk about the greatness of Martin Luther King, Jr. And then let's imagine that a Ku Klux Klansman comes up to me after my talk and says, oh, you have now hurt my feelings. What, am I supposed to be, am I supposed to apologize? Am I supposed to be regretful that my talk about the greatness of Martin Luther King, Jr. has hurt the feelings of the Klansman? No, my response is gonna be, you need to reevaluate your feelings because actually your feeling of hurt is unjustified. And no, there's no apology coming from me." }, { "from": "human", "value": "But there's a kind of line and perhaps it's a gray area. And maybe the word hurt has been overused because if you try to avoid hurting a small fraction of society that is mentally weak in a way where everything hurts them, that's the wrong way to build a society. But if we flip that upside down and say, trying to maximize the amount of love in the world and think about decisions we make in terms of language we use to try to maximize the amount of love and not just short term, but long term. And that's where freedom of speech is very powerful because it's a short term painful thing often, a long term beneficial thing, just having freedom. And so there's where the question of the N word starts to come in. How much, how do we think about its use on the internet on college campuses in a way that maximizes the amount" }, { "from": "gpt", "value": "of love and compassion and camaraderie in the world? I think that's good. And I would associate myself with your vision, how can we maximize, how can we maximize love? No, I love that, I think that's great. So let's take that on. I doubt that the way of doing that is to erase the infamous N word. I doubt that the way of doing that is to say to people, we understand that your feelings are really hurt and we're gonna do all that we can to avoid this symbolic action, maybe this word or maybe this symbol. We're gonna really, we're gonna do all that we can to suppress it so that we can have a more loving universe. I doubt that that's the way to do it. I think a better way to do it, I think a better way to do it would be to fully educate people, including educate people such that, you see that over there? That is the uniform of the Ku Klux Klan. I want you to be educated about the uniform of the Ku Klux Klan so that you can look at the uniform of the Ku Klux Klan and know what it's about. You're not terrorized by it. You're not immediately, you don't see it and sink to your knees and start wailing and crying. You don't see it and say, huh, I'm traumatized? No, you don't do any of that. You see it, you understand it, if somebody asks you about it, you're fully prepared to talk about it. Seems to me that attitude, that poise, that strength, that knowledge would be a better way of equipping us to have a more loving world. And by the way, just so, I would say that about the swastika, I would say it about the infamous N-word, I would say it about all of the things that we're talking about. It would be better for people to be educated" } ]

[ { "from": "human", "value": "so that they are not traumatized. So do you, the N-word should not be removed from Huck Finn, Adventures of Huck Finn," }, { "from": "gpt", "value": "or from the works of James Baldwin and Toni Morrison? Not at all. In fact, it seems to me that the bolderization of these great artists' literary work, as far as I'm concerned, highly objectionable," }, { "from": "human", "value": "highly objectionable. When is it okay for a white man to say the N-word" }, { "from": "gpt", "value": "with a hard R? Well, you know, here we need to focus on the word say. Is it ever okay for anyone, you know, black, white, pink, yellow, I don't care, red, orange? Is it okay for anyone to use this word in a way to put down people, to terrorize people, to intimidate people? Answer, no. But, you know, and I'd say that black person, I've seen, I mean, by the way, I've seen black people use this word to try to intimidate, put down other black people, bad. So I'm against it with a black person, white person, doesn't matter. On the other hand, on the other hand, imagine a white person who is giving a lecture on the history of American racism, a lecture on the history of American racism. And in giving that lecture, quotes the, you know, white racist politicians who until fairly recently in American life, use the infamous N word. So imagine a white history professor giving a lecture about the history of American racism and says, in 1948, this is what so-and-so running for the presidency of the United States said. And then quotes a paragraph or two in which the infamous N word is featured. Is that bad? No, that's not bad. No, that's not bad. Sounds like a perfectly good lecture. And I'm glad that you put the infamous N word in there so that we can see that as recently as 1948, people running for the presidency of the United States openly used the word. That does not bother me. I'll say this too. Now, you know, somebody says, well, you know, nice job, Kennedy, but you've, you've limited it to over here. You've put it in an academic setting. What about other settings? It does not bother me. Let's imagine somebody who's a comedian, a white comedian who is satirizing word usage. Let's imagine a white comedian who is satirizing our current practice and wants to poke fun at the way in which, let's imagine you have a black rapper who invites people on stage. And let's suppose they invite a black person on stage and they're perfectly happy when the black persons flout with their lyrics. They invite a white person on stage. The white person is, you know, that doesn't really sort of mystified, but it comes on stage. And flout with what the rapper says, including the infamous N-word, and then the black rapper gets mad. Imagine the white comedian who satirizes that, pokes fun at that, and in poking fun at that, says the infamous N-word. Am I angry? No, I'm not angry." }, { "from": "human", "value": "Not angry at all. What if in the process of satirizing" }, { "from": "gpt", "value": "that comedian is not very funny? You say that bad joke, you were not funny. I don't object to the use of the infamous N-word. I just say, you know, mm, you're not very funny, N-word?" }, { "from": "human", "value": "I just say, you know, you're not very funny. But because there's a line, when the joke is not funny, it just seems like the comedy's used as a cover to actually say something hateful. It's interesting thing about comedy. I feel like the funnier you are, the more you can get away with. Probably. And that's something to do with the thing we said earlier, which is when you use words that have power, you should do so with skill and competence" }, { "from": "gpt", "value": "and the responsibility those words carry. Again, the cases that I'm most familiar with are the cases involving teachers, professors, you know, academic, and it is said sometimes, oh, you know, why do I object? I object, sometimes people say, because, you know, I suspect that this teacher just wanted to say the word and that all of this is a cover, a pretext. Do I, you know, in my sense, my sense of it is, no, I don't think it's a pretext. If, you know, if it's pretextual, that's bad. But in my experience, I have not seen that." }, { "from": "human", "value": "I don't believe that that's what's going on. I agree with you on that, but sometimes I do see people that kind of have this fly flying towards the light, desire to say something controversial and edgy, and they don't realize that there's a responsibility there. There's a skill that you should, you shouldn't just say, I mean, actually when comedians first start out, they'll sometimes go into that territory. They'll say edgy stuff that's totally not funny. And then you realize this is not like the edgier the thing, the more skills required to really serve. Like if you're cooking as a chef, a poisonous fish, there's a responsibility on how to cook the damn thing" }, { "from": "gpt", "value": "so you don't poison the people eating it. I would agree, but to get back to your question, let's imagine that somebody produces a new set of Lenny Bruce albums. So Lenny Bruce was white, Lenny Bruce used the term nigger in his sets. Question, would I want Lenny Bruce's albums now to be purged of the infamous n-word?" }, { "from": "human", "value": "My response, absolutely not. There's another follow-on sort of question, what you think, I've seen interviews you've done about this particular book, the title of those interviews on YouTube and elsewhere, would use the full word. What do you think about that?" }, { "from": "gpt", "value": "Should I use this word in the title? In my view, the answer would be yes. Again, there are people, I've been on interviews, I've been on stage with people who had a different conclusion reached. And I respect a different way. Again, there is a place for euphemism. Again, it all depends, I'm not offended. Let's imagine that you posted this and you had asterisks instead of spelling it out. Would I be offended? No, I wouldn't be offended. I would prefer the full spelling, but I would understand where you're coming from. Would it offend me?" } ]

[ { "from": "human", "value": "No, I wouldn't offend. Well, here's the weird calculation is which version of the word in the title, as silly as that is, brings more love to the world. It's hard, and basically your answer is, I don't know." }, { "from": "gpt", "value": "Your answer is kind of, you have to do it and find out. I'm not sure, again, there's certain questions in which, there's certain questions you're not gonna, we're not really going to know, there's no sociologist who's gonna be able to tell us that. So what do we do then? Then we go to secondary positions. And my secondary position is, I don't know, one thing that I hold on to, however, very strongly, is the virtue of openness, the virtue of transparency, the virtue of freedom. And I feel as though, if I'm holding on to those things, if I'm trying to engage in a serious conversation in which I'm trying to make other people understand me and I'm listening carefully to other people, I feel at ease, I feel at ease. I feel this is going to eventuate in something positive." }, { "from": "human", "value": "And I feel okay. You are at Harvard. You're one of the most respected people in the history of Harvard. That said, you did write a book with the N-word in it, and you also have a lot of opinions that challenge the mainstream perspectives on race from all sides. Hopefully we'll get to talk about some of them. But what's your view on Harvard and universities in general and speech? Did you feel pressure from any direction on, first of all, the title of this book, the content of the book," }, { "from": "gpt", "value": "and in general, your views on race? Yeah, I am very laudatory of Harvard University I've been at Harvard since 1984. I think it is a wonderful place to work. I have, in the various positions I've taken, particularly with respect to this book or all my other books, what has Harvard University done? Harvard University has done nothing but provide support, sustenance, encouragement. I think that people get down on Harvard University. I would say to anybody, imagine the following. Imagine that the ethos of Harvard University became the governing ethos of the United States overnight. Tomorrow we would wake up in a much better United States of America. I've been supported by Harvard University. I think well of Harvard University. That's not to say that I don't have criticisms of it. But by and large, Harvard University, more than by and large, overwhelmingly it has provided me, and I think it overwhelmingly provides my colleagues with a work setting in which they can do their work without fear, and that's a good thing. Are there certain aspects of Harvard University about which I'm critical?" }, { "from": "human", "value": "Yeah, sure. By the way, I think a few people, rightfully or wrongfully, would disagree with you that if the ethos of Harvard University took over the country, it'd be a better place. But there's a lot of interesting ways to break that down because, Harvard, there's not one ethos, there's a lot of things going on that are very interesting. But one of the things that's happening is the disproportionate and aggressive growth of the administration versus faculty and students. I think the power of the university should always be with the faculty and the students. That's where the beauty is. That's where the flourishing happens, and the more you have rules, and bureaucracy, and all that kind of stuff," }, { "from": "gpt", "value": "the less powerful the university is going on that are very interesting. I think that at my university and at many universities, that's right. There's too much bureaucracy, too much regulation, and are there dangers to freedom of expression at my university and at other universities? Answer, yeah, there are, there are. So, and this has really hit home for me. There was a period of time in which I was getting off of, I'd gotten off of all boards. I was just doing my work, forget it, I'm just gonna do my work, I'm not gonna be associated with any organizations. In the last five years, that has changed quite dramatically. I have gotten on various, I've reassociated myself with various organizations, mainly organizations involving academic freedom because of what's going on on university campuses. Again, I have been, at least thus far, thus far." }, { "from": "human", "value": "This hasn't pinched me where I live, but." }, { "from": "gpt", "value": "You mean in the space of ideas. In the space of ideas, in the space of speech, in the face of teaching, I haven't been pinched, but I am concerned about things. So, for instance, let's imagine that you're applying for a job, you know, you wanna be an assistant professor. Or let's suppose that you're seeking a promotion. On many university campuses, you are asked to give a DEI statement in which you say, I plan to, you know, one of the reasons why you should hire me or one of the reasons why you should promote me is because I'm going to, you know, advance the," }, { "from": "human", "value": "you know, the DEI ambitions. Diversity, equity, and inclusion for people who don't know. This is the general, the set of programs that most universities now have." }, { "from": "gpt", "value": "Yeah, yes, that's right. So, you know, you've gotta sort of, you know, basically what you're being asked to do, whether they say it explicitly or not. They don't say this explicitly, but this is what they're, this is what is up. What you're being asked to do is to say, I'm down with the diversity, equity, and inclusion ethos, program, policy, campaign. And here's what I've done that shows that I'm down with this program and therefore I'm okay. Well, you know, a lot of what I do would fit very comfortably within that. But let's suppose that I didn't, just suppose I didn't like this. And by the way, there's certain aspects of the DEI, you know, industry that I don't like. You mean to tell me that, you know, I'm being judged at an academic institution. Let's suppose I want to be a chemist. Let's suppose I want to be a physicist. Let's suppose I want to be, I don't care, you know, a, you know, a critic of literature. I oppose this program. I don't, you know, I'm, I, I don't think this is the way in which higher education should be going. Should I have to, on pain of relinquishing my ability to be hired? Should I have to sign on to this? Just suppose, and let's change it around. Let's not make it a DEI campaign. Let's make it a Makeup Erika Great Again campaign. What would we think then? Let's suppose it was something that said, instead of it saying DEI, let's make it, let's make it say, the advancement of American capitalism as we know it. We want you to be down with that. What have you done that shows us that you believe in the advancement of capitalism in America? Would I be happy about, no, I would say this is no. Well, no with respect to these, as far as I'm concerned with the, you know, the DEI statements, oh, here's another one. I just learned, and in fact, I mean, there's certain things that are happening, and I must say, I mean, I'm in academia, but it's news to me. I didn't know until relatively recently about positionality statements. So these are statements in which somebody writes an article. Let's suppose, you know, I write an article, and it's not enough for me just to submit my article to some law review or to some, you know, some other sort of journal. No, in addition to me submitting my article, I've got to give a positionality statement in which I say whether I am, you know, gay or straight or what have you, in which I say my race, in which I say my nationality, in which I say my stance toward this ideological position" } ]

[ { "from": "human", "value": "or that ideological position. Interesting. What?" }, { "from": "gpt", "value": "Is this becoming a kind of a standard? I don't know how widespread it is. I know there was a very good article in the New York Times a couple of days ago about these positionality statements. And in fact, that's what sort of tipped me off. Somebody had told me there was a law review at my home institution. And I had a friend who sort of mentioned this offhandedly and who said, well, I submitted an article to this journal and I was a little bit taken aback in so far as they did have me fill out a questionnaire in which I was required to state my race, state this, state that, state the other. And, you know, as far as I'm concerned, well, what does that have to do? With a proper assessment of somebody's work. This concerns me. I'm concerned about the fact, you know, a little while ago, you mentioned, a little while ago, you mentioned the word Negro. And I was talking with colleagues a couple of months ago and somebody mentioned that this word had come up in their class because what had happened was one student was reading from a Supreme Court decision and the word Negro was part of what they read out. And another student held up his hand and said to the student who was reading, A, you should be careful because, you know, I find the word Negro offensive and you need to be careful about even saying a word that would be offensive to someone. And this person, and then, you know, the teacher was, you know, what should I say in those circumstances? What should I have said? And I volunteered, you know, and I said, well, guys, that's really interesting to see if that had come up in my class. I would have said, well, frankly, I don't, you know, I don't see what the, I don't even see what the big deal is because I use the word Negro. And, you know, Harvard University is not on some island that is apart from everything else that's happening in the world. If these things are happening in other places, if they're happening at Stanford, if they're happening at Yale, if they're happening at Columbia, you know, they're gonna happen at Harvard. But thus far, and I am most especially experienced in life at Harvard Law School. Harvard Law School is an open environment in which ideas are tested and they are tested fully. And it's because of that that I say I have been fully supported at Harvard Law School, feel that it is an excellent place in which to do work. I'm a fan, I am a fan, and I'm not embarrassed to say it. I am a fan of my workplace, Harvard Law School." }, { "from": "human", "value": "Very happy to be associated with Harvard Law School. Zooming out in general in education, there's something called critical race theory. Can you comment on what are your thoughts about this kind of perspective on race and race in America to the degree that it's becoming a part" }, { "from": "gpt", "value": "of the education program? Okay, so the first thing I want to say. What is it? Well, the first thing I want to say about critical race theory is that critical race theory has become a term, so I'm gonna put quotation marks around the term critical race theory. We can, I mean, in a minute I'll talk about critical race theory without quotation marks, but to begin with, I want to talk about critical race theory because the reason why people are talking about critical race theory so much now is because politicians, mainly Republican right wing politicians, have created a boogeyman, critical race theory with quotation marks around it. They have created a boogeyman and they have tried to make it seem as though this boogeyman believes all sorts of ideas that Americans should loathe and that Americans should fear. And they've created this boogeyman and they've created it, and they've done a very good job of creating the boogeyman and they have mobilized sufficient public support such that there are a number of states that have passed laws prohibiting the teaching of so-called critical race theory. Now, the first thing I wanna say about this is that this campaign, these laws, these various policies, telling teachers don't teach this and don't teach that and you can't use this book, you can't use that book, this is a frightening encroachment on freedom, freedom of speech, freedom to learn, freedom to listen, freedom to read, that's terrible. And it's one of the most frightening things that has happened in American life in recent memory. So that's the first thing I wanna say about so-called critical race theory. Now, I'll say something, I'm gonna take the quotation marks off of the term critical race theory. Critical race theory is a sort of a, you could have a nice conversation about actually what it is. One way of viewing it is to say that, well, critical race theory is a community of ideas that comes from a community of people. The community of people would be people in legal academia in the period, 1908 starting and probably the middle of the 1980s. It would be associated with people like Derrick Bell, it would be associated with people like Kimberly Crenshaw, people like Charles Lawrence, people like Richard Delgado, people like Mary Matsuda. And these are folks who held, embraced a couple of, they articulated a couple of propositions. One of their propositions was that liberal race policy was insufficient. They would say that the racial policies of a person like my old boss, Thurgood Marshall, the liberal racial policies were insufficient to grapple fully with the pervasiveness and the depth and intensity of American racism. Their basic claim, and I think by the way, it was a good claim. Their basic claim was that American racism is more central, more deeply embedded in American life than most people perceived, including liberals. And I think there was a lot of strength to that proposition. But then they also took on some other propositions with which I was in very strong disagreement. So I think it's perfectly fine to say that racism is a force in American life that is deeper, more pervasive, more stubborn, more resilient than I think people often, often understand, often perceive. But then some of the folks in critical race theory pushed further. One of the propositions that some of the people in critical race theory took was the proposition that America was doomed to always be a country that would be governed according to the dictates of white supremacy. Derrick Bell, who was a colleague of mine and a friend of mine, took that position. He talked about the permanence of racism in American life. And he took the position that the various changes that had been wrought in American life were really mainly cosmetic. They didn't amount to a whole lot. I mean, Derrick Bell took the position, the second reconstruction, the civil rights movement. Well, yeah, it made changes, but at the end of the day, black people were still, after the second reconstruction, were still in a position of almost, I don't know, some of them would even say neo-slavery. Well, I think that's ridiculous. The second reconstruction changed a lot. And as for neo-slavery, neo-slavery, what are you talking about? A black American was president of the United States between the years 2008 and 2016. I mean, what are we talking about here? There's been a tremendous change, and I think people ought to understand that now. Am I saying that everything is peachy keen and all right? No, the United States is still, to a very large extent, still a pigmentocracy, but that doesn't mean that a lot hasn't changed, a lot has. So I disagree with certain tenets of critical race theory and have been very outspoken in my disagreement. There's another one, by the way, I need to mention, because we've talked so much in our discussion about freedom of speech, freedom to teach, freedom of listening. Another big problem that I've had with some of the people who talk of themselves as critical race theory people has to do with their attitude towards freedom, freedom of speech. Some critical race theory people think that the American legal system is wrong in the latitude that it gives to what they call hate speech or the latitude that it gives to what they would view as racist beliefs. Some of the people who associate themselves with critical race theory think that racist beliefs ought to be expunged with the aid of state power, if need be. Well, I'm against that. And I think we are at a moment, an ironic moment in which actually it's the right wing that has embraced some of the ideas that were championed by some of the people who call themselves critical race theorists. They say, oh, we ought to expunge hate speech. Well, the right wing is saying this critical race theory, that's hate speech, so let's expunge it. And so I, again, I've been very outspoken in my criticism of some of the illiberal dimensions of critical race theory. So I've been a critic of certain features of critical race theory. I have applauded certain features of critical race theory. Critical race theory, there's some aspects of it that I think have been useful. There's some aspects of it that I think have been profoundly wrong at it. So that's where I am. And I certainly, and, above all, I certainly am against any efforts to remove it from the intellectual universe. It is a part of our intellectual universe. People ought to know about it, and people ought to debate it, and people ought to be free to make up their minds to conclude what they will" } ]

[ { "from": "human", "value": "about the strengths and weaknesses of critical race theory. And we'll talk about the pessimistic and the optimistic perspective on race in the history of the 20th century in America. I think you have very interesting perspectives there. But before that, I'd love to look at the current moment. And you had a conversation with Glenn Lowry and John McWhirter, and from there, it became clear to me, I think John made clear, how important to the conversation about race is policing in today's society. That that's where a lot of African Americans feel is sort of the pinnacle of racism sets the people that believe there's still racism in America. There's still a lot of racism in America. That's where it is. So to what degree do you think" }, { "from": "gpt", "value": "there's widespread institutional racism in policing? Yeah, well, my first book was a book called Race, Crime, and the Law. 1997. 1997, wow. Time flies. Time flies. Unfortunately, unfortunately, the impetus behind that book stands. That book was propelled by a sense that with respect to the administration of criminal justice, African Americans feel deeply aggrieved and they feel deeply aggrieved with good reason, and they feel deeply aggrieved with good reason in at least two dimensions. On the one hand, African Americans suffer from under protection. And in fact, in that book, the central theme of that book was that Black Americans suffer from under protection. If you take a look at the broad, sort of the broad trajectory of American history and ask yourself, in what way have Black Americans been most oppressed? Well, take a look at the antebellum period, period before the abolition of slavery, before the abolition of slavery and the locales where most Black people resided, namely the slave states. In a lot of those areas, question, was there a crime called the murder of a Black person? Answer, for a long period, the answer was no. There might've been a tort, you know, so if a white person killed the slave, a killed a slave, that person could be sued because they had injured the property of another and would have to pay money to, you know, for that. But had they committed a crime, answer, no. In the antebellum period, were Black women protected against the crime of rape? In most states, the answer was no. There was no such crime. Let's go to after, you know, slavery is abolished, thank God, slavery is abolished. Slavery is abolished. Then, let's see, you know, what happens? What, you know, so we hear lynching. Lynching, from 1890 until, let's say, 1930. Well, you know, in 1890, there was probably, you know, I would say there was probably gonna average a lynching every day in the United States. You know, well over 300 lynchings. It goes down, that was the case in the 1890s, probably the first decade of the 20th century, and then it starts going down. What was lynching about? Lynching was about Black people being executed outside the law. Did the legal system do anything about that? Answer, no. You know, show me, show me cases in which people were prosecuted criminally for engaging in lynching. You come up in most places with a null set. Black people suffered the under protection of the law. Do Black people still suffer the under protection of the law? The answer is yes, and people talk about the Kerner Commission Report, 1968. Black people were asked, you know, with respect to the police, what's your main complaint? In many places, the main complaint was, we don't have police protection. You know, when things happen to us, when our houses are burgled, when our businesses are encroached upon by robbers, when our businesses are robbed, when we're assaulted, you know, nothing happens. The police protect white people, they don't protect us. Under protection, our society right now, if you take a look at the statistics, who is most liable to be raped, robbed, the victim of assault, what have you, Black people. I mean, and it's not even close. Under protection. So that's one way in which the administration of criminal justice harms Black people by not doing what government is supposed to do, which is protect us. Thus, 14th Amendment, you know, protect, equal protection, I underline protection of the law. So that was a big theme of race, crime, and the law. Now second thing, second, and this is the thing that gets most attention, and it's important, I think, that the under protection story does not get enough attention, but then there's a second story. The second story is that Black people have historically and still today, Black people are subjects of invidious racial discrimination when it comes to, when it comes to police action. So, you know, walking down the street, walking down the street. You have a Black person who's, let's say, 20 years old. You have a White person who's 20 years old. Let's make them men. Both just walking down the street. And the question, attitude of the police towards these two. An attitude, you know, is a complicated thing. It can show itself in various ways. It can show itself in a look. It can show itself in who gets the look. You know, Black persons, you know, walking down the street or running down the street, White persons walking down the street or running down the street. What happens with respect to the police? Let's suppose that, you know, who gets the second look? Who is followed?" } ]

[ { "from": "human", "value": "Who is detained for a moment? Well, some of it is just on a small tangent. I apologize to interrupt, but attitude is an interesting one because a lot of it, a lot of the interaction doesn't show up in the data. So detained, for example, starts showing up in the data. But before then, the second look, the third look, the first look, this is where the gray area of conversation happens because- Very much so. Culture and society happens in the stuff" }, { "from": "gpt", "value": "that doesn't often show up in the data. So I tell ya, this really came home to me. Several years ago, I was in New York City and it was at a time when there was a lot of discussion over the- Stop and frisk. Stop and frisk, basically, you know, racial profiling on the street. I was walking, I was in Harlem, I'm walking down the street. And frankly, you know, the police weren't bothering me. I'm just walking. Police weren't bothering me. But you know, I'm of a certain age. I did notice though, I was looking at the police and the way that the police attitude, you know, had to do with body posture. It had to do with, that's right, who got a second look. It had to, I noticed, I'm walking down the street, I'm walking in Harlem. There, you know, there are white people in the street. You know, most of the people that were black, some Hispanic, the level of contempt, the level of animus, the level of unfriendliness that was pouring off the cop. The police, they weren't, they didn't say anything. No, they didn't detain. They didn't say anything. Bring off. It was palpable. I could feel the attitude that was being directed at the young black men. And the thing is, see, the thing is, it's not as if this doesn't matter. It matters because the way I saw it, these young black men knew, they felt the contempt that the police were shedding. And this was gonna have a consequence. The consequence it was gonna have is, let's imagine that the police did say, excuse me, you know, what are you up to? Now, if you have been feeling this contempt, if you feel like the officer who was asking this question doesn't like you, doesn't know anything about you, but just doesn't like you on site, you might answer in a certain sort of way. You're not going to give the cop the benefit of the doubt and basically think, well, you know, policeman's just asking me this. Probably just trying to make the neighborhood safe. Well, if that's your feeling, policeman's just asking me this, trying to make the neighborhood safe, well, officer, the reason why I'm here is such and such, and you know, thanks for your service. That's one response. Another response is I'm not going to tell you anything. You know, I'm not going to tell you anything. I know that you don't mean me any good. I'm not going to tell you anything. Am I free to leave? And then the cop, having heard that, then says something bad, you know, after five minutes, what do you have, you have an altercation on your hand. And I felt that, and that is part, and you're absolutely right. That's not written down. It doesn't get to court. It's there, but it's an important part of street life." }, { "from": "human", "value": "It's an informal part of street. And it has ripple effects because that young black man will probably talk shit about that cop later that day. So the narrative persists, and then the cop will also talk shit. And then there's these narratives." }, { "from": "gpt", "value": "And I think the contempt is such a powerful thing. It's so hard to disentangle because you're absolutely right. The young man, let's suppose that the story ends, quote, well, he's going to go and he's going to be talking with his friends and he's going to say, let me tell you what just happened to me. And his friends are going to say, oh yeah, that doesn't surprise me. Let me tell you what happened to me. And for two hours, this goes on. The anger, the feeling of humiliation, the feeling of aggrievement grows. It's disseminated. And that's part of what we have. But that's not all of, that's an important part of what we have, but we have, it's even worse than that because then you ask the question, what about things we do know? I know this from my teaching. This was brought up, there was a lawsuit in New York City. And notice I didn't say, I didn't say Birmingham, Alabama. I didn't say Atlanta, Georgia. I didn't say Tallahassee, Florida. I didn't say the Deep South. I didn't say Montana. I didn't say Idaho, New York City, cosmopolitan place, metropolis. In New York City, the police were challenged with respect to their policies. A judge wrote a very lengthy opinion and the facts were rolled out and the facts were really quite horrifying. People, there were black men who had been stopped many, many times. It wasn't just once, over and over and over again, under circumstances in which they ought not have been stopped. And this has real consequences. It doesn't just show up here though, of course. It also shows up in other places with respect to the administration of justice. And we still have a big problem. Now you mentioned the police. If you ask yourself, who are the state agents that are most consequential? The police. I mean, you walking down the street, what other agents have guns on them? What other agents are authorized by the law to shoot you under certain circumstances? It's the police. The police are the most consequential agents of the state" } ]

[ { "from": "human", "value": "that most people interact with. I mean, to push back a little bit of consequential in a physical sense, but if we return to the power of the psychological sense of contempt, I would say store clerks and stuff like that" }, { "from": "gpt", "value": "can also be a source of contempt. They're not agents of the state." }, { "from": "human", "value": "Right. But if we look at the landscape of contempt, which throughout the 20th century, or the bus, right, you can experience the same kind of contempt in other aspects of society." }, { "from": "gpt", "value": "But yes, the cops have consequences. In other, but yes, the cops have consequences. I would still put the police. The policeman or the police person is the person walking down the street with a gun. Gun, yeah. And if you think about the way in which the law, the extent to which the police are authorized to use their force, the police have extraordinary, extraordinary authority. Look, you're driving your car and you're speeding. The police can arrest you right then and there and take you to jail." }, { "from": "human", "value": "That's an extraordinary path. But have they also have, because of that, the leverage, just one human to another, they have more leverage to be an asshole and to show contempt to you. They certainly do. To be the lesser, to lean into the lesser aspects of their nature as all humans can just to be an asshole to show contempt. You had a bad day." }, { "from": "gpt", "value": "They have the more freedom to do that. They certainly do, yes, and that's why, that's why police officers are very important. I recognize, I know police officers. They have a very difficult job, a very important job. Very important, again, remember what I just talked about, under protection. I want the police to protect me. I want the police to protect me from the rapist, the robber. So I, you know, the police, I'm with the police. We need good policing. But we need good policing and for good policing, we need accountability. And one of the scandals, one of the just absolute scandals of American law is the extent to which the police are not held accountable. It's absolutely remarkable the degree to which American law fails to properly hold police accountable. They have an important job, a difficult job. I want them to be very well paid. And as far as I'm concerned, you know, we should be, police should make more money. They should be given the importance of what they do. They should have more respect, more prestige, more money. With all of that, they should be held accountable. And the way things are now, they're not held accountable. And every day, we see the consequences in our newspapers or just, you know," }, { "from": "human", "value": "talking with people. So what do you make of the different perspective on this? From to bring up a person that I'll probably speak with Heather McDonald, who wrote a book called War on Cops and will often bring up the stuff that does show up in the data to show the disproportionate amount of homicides committed by African Americans. And we'll also justify racial profiling on that basis in stop and frisk programs. And we'll also bring up things like the Ferguson effect saying that because of this pushback and all the stuff we've been saying about police in those areas, the police will step back" }, { "from": "gpt", "value": "and crime will increase. I was in one debate with her. And one of the things that I said is it was a debate that was sponsored by the Federalist Society, conservative legal group, conservative law students at Harvard Law School. We had a, she made a presentation and they asked me to respond. And one of the things that I said was when we're talking about the police, I'm disappointed with the reaction of some of my conservative colleagues. And I would consider her to be one of my conservative colleagues. Because what are the sort of the, what are some of the important precepts of conservatives? One very important precept of conservatives, limited government, limited government. conservatives talk about the tendency of government to overreach itself and governmental agents to overreach themselves. And I say, you're right. You're right. I just, you know, right on. Why is it that you somehow forget that when you're talking about the police? When you're talking about the police, you seem to be unaware of this tendency that is so much in your consciousness and other places. But now when you're talking about the police, you make it seem as though I'm, you know, other people are being paranoid when they talk about the danger of overreach. We need to be very careful about the danger of overreach. Another thing with respect to the police, what to conservatives, transparency in government, right? You're absolutely right. We need transparency in government. So why is it that so often when we're talking about policing, why is it that conservatives actually embrace police unions and are the enemies of transparency? Why is it that you want to prevent the citizenry from knowing that officer so-and-so has, you know, there have been 10 incidents in the last year in which citizens have complained about officer so-and-so. Don't you think that the citizenry ought to know that? Why is it that you want to keep that under wraps? So overreach, transparency, the tendency or the, you know, the problem of governmental agent corruption. Why is it that those sorts of things are forgotten about when some conservatives, by the way, not all conservatives, not all conservatives, there's some conservatives who have stuck and appropriately so to their guns and have said, hey, we need to make sure that the police stay in their lane. We need to make sure that the police do not overstep constitutional bounds. We need to be, we need to insist on transparency. There are some conservatives who have taken that line and I salute them. But there are a lot of conservatives, and I would say Heather McDonald was one of them, who all of a sudden become just, you know, totally uncritical status" } ]

[ { "from": "human", "value": "when they're talking about the police. So the overstepping, overreach, the lack of transparency. Of course, we see this kind of stuff in foreign policy as well, which is starting military conflicts and a lot of the supporters, at least with the Iraq and Afghanistan wars were conservatives and hawks and so on. So there's a lot of hypocrisy in terms of principles and so on. But maybe one question is, in this discussion about racism and policing, there was a lot of cops that might listen to this and feel like they're not being, their profession is not being respected. They're not being heard. Not being respected to the difficulty of the problem they're facing." }, { "from": "gpt", "value": "Can I just interject? Remember a few moments ago. I mean, if I were in charge of things, if I were in charge of things, I'd pay police more, in fact, much more than they receive. I do, police have a very difficult job, extraordinarily difficult job. I mean, for one thing, sort of maintaining the law. The law, that's a very complicated thing, the law. That's not an easy thing. I mean, I teach law. I spend a lot of hours with very smart people trying to understand the law, very difficult. So here we expect people to understand the law at the same time that they are grappling with people, some of whom are violent. It's very difficult. I respect police officers. And I said, we need police officers. You did not hear me say, by the way, so let's get down to brass tacks. You did not hear me say, defund the police, did you? In fact, to the contrary, I said, defund, actually I want more funds for the police. I want to hold them accountable. I understand the difficulty of their job. I respect the police. I want people to respect the police. I also want the police to respect civilians and demand that they do. I am not a police abolitionist, far from it, far from it. Again, we need good policing." }, { "from": "human", "value": "For good policing, we need good people. To what degree, almost philosophically, maybe practically, do you think police should be doing racial profiling?" }, { "from": "gpt", "value": "Profiling in general." }, { "from": "human", "value": "Okay. It's a very difficult philosophical, moral human question." }, { "from": "gpt", "value": "Yeah, so first thing we need to do, and here we get back to the early part of our conversation because we focus so much on, first part, a word. Here we have another word, profiling. Question, what is racial profiling? Now, here we have, in the weeds, we have people talking past one another. I've been in conversations with police officers and debates, and we talk past one another because we are defining racial profiling very differently. If you ask, I've been in conversations with police officers and they say, oh, I'm totally against racial profiling. Then I say, well, sir, what do you mean by racial profiling? And here's what they say. Racial profiling is when police officers act against somebody wholly on the basis of race that gets rid of the issue because most police officers don't act against anyone wholly on the basis of race. You could have the most racist police officer, and that police officer is not going to act adversely against the 90-year-old black woman walking down the street with a cane, okay? So it's always, you know, if you define racial profiling that way, you're getting rid of the issue. The issue is, the proper issue is this. Should the police be able to act against someone taking race into account as a factor? So let's imagine that a black person who's 25 years old is walking down the street right next to a white person who's 25 years old. Some people would say, you know what? Under those circumstances, it's okay for the police officer to give two looks at the black person and only one look at the white person. Why? Because the statistics tell us that the black 25-year-old, there's much more risk of that person acting in an unlawful way. If you take a look at, you know, crime statistics and ask the question, with respect to, you know, homicide, with respect to robbery, with respect to various, you know, crimes, is there, is there a difference between the white 25-year-old and the black 25-year-old? With respect to certain crimes, the answer is, in certain places, the answer is, yeah, there is a difference. And there is a greater risk that the black 25-year-old has engaged in various forms of criminality. I'm not, you know, that, okay. Under those circumstances, what do I say? Should the police officer, therefore, be allowed to take action vis-a-vis the black person, as opposed, you know, as against the white person? My answer is no. I'm fully willing to concede, let's concede for the point of, you know, for the point of discussion, that there is more risk. There might be, there might be. But, for reasons of constructing the sort of society I want, we should not empower agents of the state to act towards certain people in a way that's, you know, adverse to them. Let me try to break this down in a little different way. Let's go back to, let's talk about what happens when you wanna get on an airplane. You know, wasn't so long, you know, 9-11 wasn't all that long ago. It was a while ago, but it wasn't all that long ago. And this very issue came up with the respect of the profiling of Muslims. And there were some people who said, well, geez, you know, in the aftermath of 9-11, profiling of Muslims makes sense. You know, no hard feelings, but, you know, the 11 people who, you know, were on those planes, were all Muslims. Now, there is a rationale. You could just simply say, it's not prejudice, it's not animus, it's not invidious, it's just, you know, the facts take us this way. Well, the facts never take us this way. There are facts, and then there is our choice of how we want to respond to those facts. Now, you could respond by saying, well, we perceive Muslims to be more likely to, you know, do bad things on an airliner. You could respond in that way. On the other hand, you could say, no, we want a society in which people do not have to grapple with prejudice on the basis of their religion or on the basis of their race. And to deal with that, we are gonna demand, we're gonna demand that agents of the state act towards people in the same way. And so if that means that everybody getting on the airplane, before they get on the airplane, has to, you know, open up their luggage and it slows things down, but everybody has to open up their luggage. I would prefer that over a system in which we focus on people who are Muslim. And one of the reasons, and I'm gonna bring this back to the racial thing. One of the reasons why I wanna insist upon that is because if, you know, if we all have to open up our luggage, what that means is we're all paying a tax for more security. And we're all likely to ask ourselves, hmm, do we wanna pay this tax? Whereas if we focus simply on the Muslims and allow the Muslims to be the only ones who are sort of paying the tax, we'll foist that on them. Same in the, you know, black person, white person, 25 years old, walking down the street. Do we want to impose a racial tax on the black 25 year old? Some people would say, yes. I say, that's a mistake. Black people aren't stupid. They know that they're paying this tax. It is a violation of, it seems to me, the rules that we ought to want to have to govern our society. I say, spread the costs, make the police deal with everybody the same, and don't allow a situation to develop in which a group of people can accurately say, we are paying more of a cost" } ]

[ { "from": "human", "value": "than these other people over here. So it's a better avoiding profile while it may have costs on security in the short term. In the long term, it's the embodiment of the principles that all men are created equal. So it has a much bigger benefit in representing the fairness that is at the core of the ideals of this country. Yes. Let me ask a big philosophical question. Why do you think there is racism in the world?" }, { "from": "gpt", "value": "Yes. Oh, now you've, okay, fine. Now you've asked me a question" }, { "from": "human", "value": "in which I throw up my hands and I don't know. Will there always be? Why is there tribalism? Are humans always finding this way to divide ourselves?" }, { "from": "gpt", "value": "And how bad of a problem is that? It's a huge problem. I don't, you know, I mean, I don't, I'm not gonna pretend that I know enough to grapple in a satisfactory way with such a question. I mean, to really grapple with that question, one really has to have all, you know, a very broad knowledge base in which they can make comparisons. I mean, the world's a big place. Now from the little bit I know about the world, and I read, I read as much as I can. I've been, I'm a very privileged character. I mean, I live in a university. So I'm reading constantly and listening constantly. From what I gather, the problem of divided societies is a problem that is worldwide. Very broad, there's, you know, and it seems as though, it seems as though human ingenuity is such that humans will find something to divide over. You know, it might be texture of hair. It might be, you know, complexion of skin. Politically, ideology. Political ideology if the texture, if the texture of skin is the same. Politically, ideology, political, it'll then be, well, you know, you wear these sorts of clothes. You speak in this sort of way. You believe this, whereas I believe that. I mean, it seems like a human ingenuity is such, they will, humans will find some way of distinguishing themselves, and then they seem to want to embellish the distinction. It's not just a distinction. It's not just, you believe this and I believe this. Okay, that makes the world interesting. Let's talk about what you believe and what I believe. No, it's usually associated with you believe this. I believe this. Of course, my belief is superior to your belief, and I want to put you down." }, { "from": "human", "value": "And now we're headed towards war. The interesting thing is, if I were to step outside of this whole thing, from an alien perspective visiting Earth, I think America is one of the greatest, if not the greatest countries in the history of human civilization. And I think that the line between white and black, the racial struggle is the thing that in part made it a great nation. There's something about the division and the way you alleviate that division through the struggle for human rights that makes for a great nation. And so it's interesting that the division is almost the fuel for the greatness, for our discovery of what it means to be human, or what justice means of, and it's interesting that it seems like there's this struggle that you're elucidating now, but that struggle itself is our search," }, { "from": "gpt", "value": "a man's search for meaning and justice and freedom. I think there's something to that as you were speaking. I was immediately thinking of a person who most jumped to mind was the great Frederick Douglass. So I mean, here you have a person who in my view is one of the great people in the history of the world. And one of the things that ironically enabled him to become one of the great people of the world. He was born into slavery. His overcoming was the very thing that in a sense made him this larger than life figure. And I think there's something to it. I will say this because a minute ago you mentioned optimism and pessimism. And I'd say in the last few years, my feeling about the United States has changed and it's changed in a somber way. For most of my life, for most of my life, I've been in the optimistic camp with respect to my view of American race relations. The optimistic camp, that was the camp. That's the camp that believes we shall overcome. That's Martin Luther King's camp. Martin Luther King, you know, Martin Luther King, hours before he's killed. I've been to the mountain top. I might not get there with you, but I've glimpsed the promised land, optimism. Frederick Douglass would be in that tradition. There would be other people." }, { "from": "human", "value": "Wonderful, good people who would be in that tradition. Would you sort of interrupt put in the pessimistic camp," }, { "from": "gpt", "value": "would you put Malcolm X on it? On the pessimistic camp, the pessimistic camp is the more interesting camp. I mean it is, it's the more interesting camp because in the pessimistic camp you have, I mean, if I was going to list some of my pessimists, pessimists, number one, Thomas Jefferson, Thomas Jefferson, the author, you know, principal author declaration of independence he was also the author of, Notes on the State of Virginia. And he said in notes on the State of Virginia, basically we shall not overcome. He talked to, he said, you know, the black people will always know that their forebears were enslaved and they will always be resentful of that, always be aggrieved by that. Jefferson did not think that we would ever have in the United States a multiracial democracy. He was very critical of slavery. Now he was a hypocrite. He had slaves, he sold slaves. He was terrible in that way, but he did understand that slavery was horrible, but he did not want, he did not want to free the slaves for a variety of reasons. But in one reason was because he thought that it would be impossible to have a society in which blacks and whites were equal neighbors. He was thoroughly pessimistic. Another pessimist, Alexis de Tocqueville, thoroughly pessimistic. So who were some of the other pessimists? Abraham Lincoln, pessimistic. That's why he was so interested in colonization. He basically said, you know, I don't like slavery, but blacks and whites are not gonna be able to share the United States. Maybe the best we can do is just, you know," } ]

[ { "from": "human", "value": "put blacks, ship blacks someplace else. Can you elaborate on that? I think people would be surprised to hear that you will put Abraham Lincoln in the pessimist camp." }, { "from": "gpt", "value": "Oh yeah. Oh, Abraham Lincoln's thoroughly pessimistic. He was, he believed, he was anti-slavery, but he did not believe that blacks and whites would be able to share the United States together. And he was always very interested therefore in colonization. Even after the, even during the civil war, he was interested in, you know, maybe all black people, would all black people be interested in maybe going to Panama? Would they be interested in going someplace else? Because, you know, Lincoln was aware of how racist white people were, including himself, or including, and he did not think that black people and white people would be able to share the United States. Now, there's a black nationalist tradition. You mentioned Malcolm X. Well, before Malcolm X, you know, Marcus Garvey. My father, my father was a thoroughgoing pessimist. His view was A, the United States was born as a white man's country. It's gonna remain a white man's country. And that's the way it is." }, { "from": "human", "value": "He was thoroughly pessimist. You mentioned just a brief aside about your dad that when you moved to South Carolina, to Washington, DC, and you asked him why, his response to you was, because either a white man was going to kill me or I was going to kill a white man. Yes. So he saw race as an important line" }, { "from": "gpt", "value": "that divided people in the United States. He certainly did. And he thought that the line... My father did not. My father had passed away before Obama was elected." }, { "from": "human", "value": "I would have loved to have talked with my dad. What do you think he would've said? And you wrote a book about the Obama presidency." }, { "from": "gpt", "value": "I did, I did, I did. And I never... What would my father have said? My father would have been delighted. My father would have been happy about his election. I do think, though... I do think that my father would have said, I'm happy that Barack Obama has been elected. Hold on to your seats." }, { "from": "human", "value": "Let's see how the white people respond." }, { "from": "gpt", "value": "So he would've predicted, Donald. I think that my father, I think that in 2016, I think that in 2016... You would've said, I told you. Yes, yes! I think that in 2016, my father would've said, all of you people who were talking about, I don't know what's happened to America, and how could this have happened, and you know, no. I think my father would've said, this is America being America, and what has happened is that America has been put off kilter by a black family being in the White House. It has deranged millions of white people, and now this is coming home to Ruth. So I think my father would've absolutely said in 2016," }, { "from": "human", "value": "I told you so. In 2016, as you're saying, is the reason that you have at least dipped your toe" }, { "from": "gpt", "value": "outside of the Optimus camp into the pessimist camp. I'm, yeah, I wanna be careful here, and I don't wanna-" }, { "from": "human", "value": "Don't give up on the Optimus camp." }, { "from": "gpt", "value": "I don't, I don't wanna, I don't. You're right, I don't wanna give up on the optimist camp. I don't, I have been, my optimism has definitely been dampened, has definitely been tested. I am certainly, no, I am not, I am not as triumphalist as I once was. I am not as indignant as I once was in my criticism of pessimists, but at this moment, as we speak, you know, am I more in the optimist camp than I'm in the pessimist camp? I'm still probably more in the optimist camp, but my optimism has been dampened. But as I speak, I have to say the following things, and I'd say the following things with my father, but again, whom I revere, great man. But if my father was sitting here with us, I would say, listen, Pop, it's absolutely true that the country is more racist than I had thought. That is true. It's also true, Pop, that the Vice President of the United States, as we speak, is a black woman. As we speak, it's true that the Secretary of Defense, now, Secretary of Defense, the Secretary of Defense is the head of the Pentagon. The Secretary of Defense knows where the button is, okay. This is not a little, you know, this is not a small out of the way thing. This is the Secretary of Defense is a black man. There are a slew of black generals. There have been black secretaries of state. There are black people who are the heads of police forces. There are black mayors. There are black people who are the heads of, you know, some of our foremost foundations. There are, you know, the president-elect of Harvard University is a black woman and we could go on and on and on and on. This is not, you know, we're not, when I was growing up, when I was growing up, when I was, let's say 10 years old, we got a magazine, Ebony Magazine, every month, and in Ebony Magazine, you could turn to the middle of Ebony and they would have Black firsts. The first black person to do this, the first black person to do this. You could read Ebony, and frankly, I could tell you all of those, I knew their names. Now, I read stuff, you know, the leading cadet, the chief cadet at the United States Naval Academy. I'm reading, I'm reading, then I see a picture. Black woman, I said, I'm reading, I didn't know that. Back in the day, I would have known that. Now, has the United States changed? Yes. Is there racism? Yes. Is it, you know, a substantial force in American life? Regrettably, tragically, yes. Has the United States changed? Yes. And again, if we want to go international, the United States is not the only country that is a country that has wrestled with deep division. You think about, I don't know, think about India. Think about, you know, the United Kingdom. Think about practically any large nation state. They've all grappled with divisions. If one asks about the United States and the race question, and one puts on the table that, you know, in 1865, now, you know, when I talk with students, I say 1865, they think, oh my God, you know, isn't that when dinosaurs roamed? No, it wasn't when dinosaurs roamed. 1865, frankly, is not all that long ago. In 1865, the great mass of black people in the United States had recently been released from chattel slavery. The great mass of black people in the United States in 1865 were illiterate. Now, I mean, it's absolutely, it's an absolutely extraordinary story. And so one of the difficulties I have at this moment is wrapping my head around two stories that are in such tension with one another. One is the continuing story of racism, which is an awful story, but the other story is a story that is encapsulated in the title of a great book of history by John Hope Franklin, From Slavery to Freedom. And, you know, those are two stories in American life and it takes an awful lot." } ]

[ { "from": "human", "value": "Put your mind around both stories and I'm trying to. Well, as an optimist, I think you put more value to the overcoming side of the story, the overcoming of hardship, the overcoming of slavery, of discrimination. Well, since you mentioned illiteracy from, what is it, 150 years ago, let me ask you about affirmative action. You wrote a book on the topic titled For Discrimination, Race Affirmative Action and the Law. First, what is affirmative action? And maybe we could talk about what your view is on it today." }, { "from": "gpt", "value": "Sure. Well, affirmative action refers to, racial affirmative action, because there's various sorts of affirmative action. I mean, you can have affirmative action for veterans, you can have affirmative action for in-state residents. Let's suppose you have a school, you know, the University of, you know, blah, blah, blah. And we're gonna, we want to reach a, we're gonna give a, reach out a hand to boost help people who live within this state. We're gonna have a boost for veterans. We're gonna have a boost for, you know, for women. We're gonna have a boost for men, in certain circumstances. Well, we're the most, the type of affirmative action that gets most attention, and that's worth noting as well, there's lots of, there's some sorts of affirmative action. But when you say affirmative action, the type of affirmative action that," }, { "from": "human", "value": "immediately people are thinking about is racial affirmative. Seems like race is at the core of this American experiment, part of in every part of it's-" }, { "from": "gpt", "value": "Bingo. Bingo. Bingo. Why is it that we, again, there's various ways in which institutions reach out to help various sectors of our population. Why is it that this is the one that generates all of the pulling out of hair and gnashing of teeth? We can put that to the side for a moment. Racial affirmative action refers to efforts in which institutions, reach out to provide assistance to racial minorities. Now the reaching out can happen in different ways. A light form of affirmative action might be reaching out in terms of recruiting, making a special effort to make sure that young people, let's say in racial minority neighborhoods, know about your college, recruitment. On the other hand, racial affirmative action might take a stronger form, and might take the form of saying, okay, we have a competition. And we think that there will be too few racial minority kids who do well enough in this competition to be admitted to our school. So what we're gonna do is we are going to, in various ways, give a boost to the minority kids. If it comes down to you have two kids, they both, I don't know, have 99 on the test. There's one place left. We only have one spot left, two kids. One's a black kid, one's a white kid, they both got 99, we're gonna give it to the black kid. Why? Well, black kid, all there is to it. Black kid, we wanna maybe, and then our theory, maybe our theory is that black people have been done wrong in the United States, and we want our institution to contribute to making amends. That might be a theory, another theory might be, we think that we'll have a more interesting student body, we'll have a better student body, a better discussion if there are more black kids on campus. And unless we make a special effort to bring more black kids on campus, our student body will be overwhelmingly monoracial. We'll just have white kids, few Asian kids, a few Latino kids, if there are no black kids here, I don't know, it'll be lacking an important aspect of American life. So that's some sort of the so-called diversity story. So anyway, there are various theories. Some are grounded, like I said, in reparative justice. Some might be grounded in distributive justice. So for instance, when people say nowadays, we want a cabinet, we want a team, we want a student body that looks like America. That might be a distributive justice theory for why you want affirmative action. Unless we reach out and give a boost to certain groups, they won't be here. And we want a campus that looks like America, and we won't have a campus that looks like America unless we give a special boost to, I don't know, the Latino kids or unless we give a special boost to black applicants. So, reparative justice, distributive justice, or third, diversity. We just wanna have an interesting student body, and to have an interesting student body, we want kids from a wide array of places. We think that that'll make for a better campus, more interesting campus. Those are three justifications, all of which, however, are justifications for, in a competition, giving a special boost" }, { "from": "human", "value": "to some people based on their race. So what's your sense? Can you make the case for affirmative action," }, { "from": "gpt", "value": "and can you make the case against it? Yeah, I can make the case for, and I can make the case against. The case for, I won't spend much time on this. Like I said, there are three main justifications. Your audience should know that the justification that actually led to affirmative action is a justification that you don't hear about in the courts, because the courts have said that that justification is insufficient. So the real justification behind affirmative action was in the late 1960s, in the aftermath of the Civil Rights Revolution, in the aftermath of the Civil Rights Revolution and the aftermath of the Civil Rights Movement. There was a feeling that well, okay, fine, we're not discriminating against black people anymore, but black people have been disadvantaged by the discrimination that has been put upon them. And this discrimination that was put upon them has disabled them. And so they go into competitions and it's unfair. Yeah, they score less well on the standardized test, but surprise, surprise, what the heck? They went to schools in which they got the leavings of the white folks. You know, the white folks gave the black folks the old textbooks. That's what the black kids read. The white kids read the newest textbooks, on and on and on. So, reparative justice. We're gonna try to repair the scars left by past racial injustice. And we're gonna, and so it's sort of an effort to overcome the vestiges of past misdeeds. That was one justification for affirmative action. And frankly, I think that's the justification that has always been the predominant justification, whether people owned up to it or not. But that's one justification. The distributive justice justification is there. Like I say, you know, people, we want an integrated America. We want an America that looks like America. In that justification, legitimacy is sometimes mentioned. So for instance, people will say things like, well, you know, if we're gonna, in the armed forces, you can't have an armed forces in which the people on the ground are, you have a lot of people on the ground who are people of color, but none of the people calling the shots, none of the generals, none of the may, you know, colonels are people of color. No, and then the people on the ground are not gonna stand for that. So for purposes of legitimation, we need to, for purposes of buy-in, we need to have a situation in which people get the sense that even if they're sort of low-down, they're still part of the story, they're part of the team. So that's part of the sort of, you know," } ]

[ { "from": "human", "value": "I dunno, distributive justice notion. Do you think there's power to that? I mean, do you think there's value," }, { "from": "gpt", "value": "there's correctness to that kind of idea?" }, { "from": "human", "value": "I think there's some. Because for the application process for different universities, I think there's probably a driving narrative, justification for, well, affirmative action, I don't know where we put DEI efforts, because they overlap," }, { "from": "gpt", "value": "but they're not perfectly overlapping. The reason that the school, so yeah, as you know, there is a case, we'll know about it, it'll be announced sometime in the next month, well, month, month and a half. It's a case of the Supreme Court of the United States in which the affirmative action program at Harvard and the affirmative action program at the University of North Carolina are being challenged. Most people think, including me, that the Supreme Court of the United States is going to limit, if not totally seek to abolish affirmative action. So this is a very burning issue. As things currently stand, my university and other universities embrace affirmative action on the third ground, the diversity ground. And what they say is, we need affirmative action because affirmative action is good for pedagogical reasons. Now, frankly, do I think there's something to that? I think there's a little something to that with respect to some subjects. I don't think that's what's really going on for the most part. I mean, if that's what's really going on, why don't we have more foreign students? Why isn't there a greater effort to bring in more foreign students? Why isn't there a greater effort to bring in more, I don't know, religious fundamentalists? Why isn't there a greater effort to bring in, and you can just name various groups" }, { "from": "human", "value": "that you hardly ever see them on campus. So you don't think this effort" }, { "from": "gpt", "value": "of driving towards diversity is rigorous enough? I think it's often pretextual. I think, frankly, the real reason has to do with the belief, and I think it's a good belief. I associate myself with the belief. It's just that because of the legal rules, the authorities can't say this belief out loud, but it's still the case that a lot of institutions want to help American society overcome its racial past. That's the real animating force. Now this thing about, you know, we'll have better classroom discussions, you know, some discussions. I mean, you know, with some subjects, yeah, listen, I'm an ignoramus when it comes to physics. My sense of it is, however, that if you're talking about physics, you're talking about physics, and frankly, it doesn't much matter in terms of mastering physics, what the demographics of that classroom are." }, { "from": "human", "value": "You either know physics or you don't. The interesting thing I've noticed since doing this podcast is one way it does matter, it's really lonely when you do a thing and you don't see somebody that looks like you doing that thing. And it's such a seemingly stupid thing. Why does it matter if there's another person that looks like you in this very shallow sense? But it seems to matter to people. And when I talk to, for example, women on this podcast, a lot of women reach out saying how inspiring that is. That's interesting, right? Like, I think we're still human and we see people that look like us and this kind of narrow, shallow definition of diversity" }, { "from": "gpt", "value": "still matters, I guess a lot. Okay, well, you know, I think you've made a good point and maybe so. So let's go back to the physics. I mean, let's suppose that under one scenario, there's a classroom and a kid sticks his head, a black kid sticks his head in the classroom and doesn't see anybody else black there and has to make a quick decision and says, yeah, you know, nah, this ain't for me versus they stick their head in the classroom. They see a smattering of other black people in the classroom. They say, oh, let me try this. And then they try it. And it turns out, damn it, they're Albert Einstein. So, you know, to that extent, I can see, you know, under that scenario, maybe it does matter. So, you know, I'll have to recalibrate. Still, those are the three, those are the three leading justifications for affirmative action. You asked me, where do I stay with respect to affirmative action? Oh, no. Before we get to that, are there criticisms? The answer is, yes. There are criticisms of affirmative action. Affirmative action is a policy like, you know, any policy. Any policy is going to have some downsides to it. And affirmative action's no different. Does affirmative action have some downsides? Yes, it does have some downsides. What are they? Well, let me mention a couple. One, stigma. Stigma. So if you have an institution that says that it reaches out to give a boost to certain people on a racial basis, there are going to be people who observe that and who are going to be thinking to themselves, hmm, if so-and-so is here and they were given a special boost, doesn't that mean that they were not as proficient as the other people who were here and who did not have that boost? So if they are not as proficient, that means that they might not be as good. They might not be as up to snuff. Is that part of our affirmative action world? Yes, it's part of our affirmative action world. Yes. I'm a professor at Harvard Law School. One of the subjects I teach is contracts. Actually, that's probably the subject I most enjoy teaching. I think I most enjoy writing about racial conflict and the legal system. But as far as teaching, I most enjoy teaching contracts. Well, on that very first day, on the very first day, when I, you know, the 80 students come, especially the first year students, they're nervous, I show up. I'm quite sure that there's some students there who are thinking, okay, well, is this guy as good as my other teachers? Because, you know, I know that, you know, institutions like Harvard have made a special effort to bring in people like this guy, Kennedy. Is he as good? Well, I mean, I'm sorry. If you have an affirmative action regime, people are gonna think that. And are they crazy to think that? No, they're not crazy to think that. What do you mean? No. That's a perfectly logical thing for somebody to think. Does it have an effect on me as the teacher? Yeah, actually, does have an effect on me, sure. I'm aware that some people are thinking that. It doesn't, I'm not, you know, it doesn't, you know, make me shake in my boots. Am I aware of it? Yeah, I'm aware of it. Does it have an effect? Yeah, it probably does have an effect. It probably does. It probably, it probably makes me, it probably makes me redouble my efforts because I don't want anybody to think justifiably that I'm less able than my colleagues. So, you know, it's funny. When I was growing up, my father used to tell me over and over again, Randy, tie, tie, you lose. Okay, brother, let me just tell you that. And he was telling me that because when he was, you know, in an earlier time, tie, tie, you lose, it was, you know, if it's even and there's a choice to be made, the white person is gonna get the benefit of the doubt. Now under affirmative action, there's been a switcharoo. Oftentimes it's the black person who gets the benefit of the doubt, but I'm still in a situation which- If to work hard." } ]

[ { "from": "human", "value": "Yeah, which. If to work hard." }, { "from": "gpt", "value": "Yes, yes. To avoid the stigmatization that is imposed by affirmative action. But is that stigma there? Yes, it's there. It is there. I'll tell you another way in which it comes up from a student's point of view. I'll never forget this. It was in my second year of law school. It was in tax class. We had a very famous tax teacher, wonderful tax teacher, on the very first day of class, Professor Bitker, wonderful man, wonderful teacher, Boris Bitker, called on. The very first person he called on was a black student. I remember this as if it were yesterday. Called on a black student, and it was a black woman student. And I remember when he called on her, I remember just I felt as if the room got quieter, really. I felt as if the room got quieter, and I know there was a real tenseness within me. He didn't call on me, because he was calling on somebody else. Was a black woman student and I felt different. I felt as if I was, I felt as if I was somehow at issue. I felt as if my place, I felt as if my status was some small degree at issue. In the positive direction of the negative. Just, at issue. At issue." }, { "from": "human", "value": "And so, now all of this is." }, { "from": "gpt", "value": "At issue. It's happening really quickly. He calls on the student. This is, you know, happening like, in seconds. Yeah. And she responded. And she responded really strongly. It was clear she, he asked her a question. She answered the question. She answered the question, you know, beautiful. I mean, you know, very strong, comprehensive, wonderful response. Now you might say, okay, well, you know, smart student. Okay, fine. Next. After class, after class, I went up to her. And I clearly remember this. I said, that was great. Thank you. Why? Because a black student, predominantly white institution, Yale Law School. And I felt some of the, again, you know, affirmative action. I felt some of the burden of this affirmative action stigma." }, { "from": "human", "value": "So it was almost like a thank you for showing" }, { "from": "gpt", "value": "that we belong here. Yes. Perfect. Boom. Yes. That's exactly what the situation. And by the way, I looked, I wasn't the only one. It, well, I wasn't being idiosyncratic. I wasn't the only one. There weren't me black people in that class, but per all of us. And we laughed about it later. We laughed about it, but that was there, there. So there's, there's the stigma issue. And some people have made a lot of this. Some people have really made a lot of the stigma issue. My attitude towards a stigma issue. Yeah, it's there. It's, you know, again, it's, it's part of, you know, it's part of the situation, but I think that the benefits outweigh the burdens, but is that a burden? Yeah, that's a burden. What are some others? Resentment, resentment in society. And that's a, you know, so in these, in the fancy schools, and by the way, you know, remember when we're talking about affirmative action, most colleges and universities in the United States, there, there is no affirmative action issue because they're not selective. Hell, they'll take anybody who, you know, if you can pay, come on in. It's only a fairly small set of schools that are selective. But of course, those are the most elite schools. Those are the schools that people most want to get into. And that's one of the reasons why we have all this, you know, fighting over those schools. Now, you know, there's a whole, how many millions of people are there in America who have applied to various places? They didn't get in. And what do they say? The white person who applied to Harvard, applied to Yale, applied to Columbia, you know, applied to Georgetown, applied to NYU, you know, didn't get in. What did they say? I would have gotten it hadn't been for that racial affirmative action. They're resentful. Well, you know, should they be resentful? No, they shouldn't be resentful, but they are resentful. Does that have a consequence? Yeah, it has a consequence. It has a consequence on how they act towards other people. It has a consequence on how they vote. It has a consequence. And, you know, that's, that's something. Um, are there other things? Yeah. And, you know, in my book, I, I have a, you know, I go through various, um, I think that, uh, I think that there is a, I think that there is a certain sort of denialism that has accompanied the affirmative action debate. So because, you know, black people like myself want to avoid the stigmatizing burden of affirmative action. There's some people who to deal with that have said, Oh, actually there is no difference between, you know, the beneficiaries of affirmative action and the other kids, there is no difference. Um, if, if the affirmative action kids got, let's say a 500 and the other kids got a 750, uh, that doesn't make any difference. That's just BS there. You know, the only reason the kid who, you know, the 550 kid didn't get 750 is because the test asks all sorts of, you know, biased, culturally biased things like, you know, what is a yacht? Well, sorry. Uh, yes, there is a difference. There's a 200 point difference. Does the difference matter? Yes. The difference matters. And by the way, if you don't know what a yacht is, you can know what a yacht is without owning a yacht. Okay. And there's this denialism that I think has really seeped into not just the, you know, and into various conversations. That can have a slippery slope effect as beyond just this, this. Yes. And, you know, we see it in various ways. So, you know, this, this, you know, this attack on testing. It's not like I'm holding the, you know, uh, you know, some tests are not good tests. I think we should be skeptical of everything, but there are a lot of tests that yeah, they tell us something. All right. They tell us who knows what other, um, there's some people who are really, you know, dead set against testing because they're dead set against anything that might show a gap. Well, I think by the way, you know, are there gaps? Yeah, there are gaps. And what we need to do is be cognizant of those gaps and do things to make it so that if there's a gap, if you're, if you're deficient, no, no shame in being deficient, heck I'm deficient about a lot of things let's, you know, let's learn so that I catch up, but others, you know, but I think sometimes there've" } ]

[ { "from": "human", "value": "been people who've been afraid of even acknowledging the gaps. Of course, there's, uh, I guess a colleague of yours, Michael Sandel, with, uh, with tyranny and merit, there's, uh, interesting, rigorous ways to kind of challenge ways on the flip side of that where obsessing with merit can go wrong also." }, { "from": "gpt", "value": "Yes. Uh, you mentioned Michael Sandel. He's a, uh, a wonderful colleague and he's a wonderful friend. I've known him a long time. I think he makes very important, uh, arguments, uh, about, uh, meritocracy. Um, I disagree with, uh, some of the points that he makes about meritocracy. You lean towards the, the importance of meritocracy. I think that, uh, yes. I think that, I think that there are values and meritocracy extremely important that in fact, you know, the, the, the movement from, you know, feudalism, the movement from status, the, the, the, the idea of, you know, I don't care who your father or mother was. I don't care from what part of town you come from. I don't care what your last name is. I don't care what your skull or of your skin is. Show me what you can do. And then somebody sits down. Ok, I'll show you what you can do. You know what I can do and they show it you're in. I think there's a lot to that and you know, the impulse, the sentiments behind that I resonate with that. I think that there's a lot there. What I want to do is I want to get rid of those features in society that deprive people of the, what you need to develop yourself, you know, sometimes those are psychological. Sometimes that's you know, you're not around people who've done things that give you the idea that you can do things. I want to get rid of that, but the idea of, you know, people. Um, the idea, by the way of a- uh, you know, distinguishing, you know, this is excellent and the people who are excellent, you know, they're here. And then there are people who are, are they excellent? No, they're not excellent. They're good. Uh, I wouldn't, I wanna, I'm not gonna, you know, close my" }, { "from": "human", "value": "eyes to that distinction. And, uh, uh, highlight that distinction and yet at the same time, maintain a sense that, uh, their basic worth as a human being is equal." }, { "from": "gpt", "value": "Yes. So, uh, you know, uh, was it run BMC? What are the rap groups? No, you know, we're all written down on the same list. We're all written down on the same list. Yes. We're, you know, uh, so I want to recognize our fundamental humanity. Um, I don't think that that, I think that one can recognize our fundamental humanity and one can also recognize as far as I'm concerned that, um, we all collectively should make sure that we, we do all that we can to prevent people from sinking below a certain level and being, you know, in misery, um, you know, all for that, but I want to be careful about some of the attacks that I hear on meritocracy. So some of the people, including again, I have all the respect in the world for, for, for, for, for, for Michael Sandelli to, you know, it talks about the, the, the arrogance of the winners. Okay. I want to be, you know, I, I don't want the winners to be arrogant. That's right. Luck has a lot to do with things. You know, you were, you, you didn't have any control over the circumstances in which you were born in two. You were lucky that you were born healthy and that you were born with a, you know, well-working mind. You didn't have anything to do with that. That was pure luck. There's some people who don't have that walk. So don't, you know, you know, don't be, you know, Mr. Big stuff, okay. So I'm, you know, I'm against that sort of arrogance. You know, I'm entitled as if, as if I taught myself how to read. No, you didn't teach yourself how to read. There was people who did all sorts of things for you and you don't even know it. Okay. So, you know, I want to, you know, get rid of the arrogance, uh, have, have decent humility on all for that at the same time, you know, um, I want to be careful about the problem of envy. I want to be prob, I want to be careful about the problem of resentment. I want to be, I want to be careful about, you know, I've heard, you know. So let's not have, let's not give a trophy to the person who wins the race. Because to give a trophy to, to the person who wins the race will make the person who they defeated feel bad. No, no, no. No, no, no, I don't want that. I want to give a trophy to the person who wins the race because I think it's a good thing to valorize the best." }, { "from": "human", "value": "Yeah. It's a, it's a kind of celebration of, um, this whole human project that we're on. Yes. It's celebrating the best. You mentioned your father several times, so let me just linger on that. What, uh, what have you learned from, what have you learned about life from your father?" }, { "from": "gpt", "value": "There's, I've been just such a lucky person. I mean, I, I feel like I've just lived an absolutely charmed life. And I live, I mean, the work that I do is what I love doing. Um, I would pay to do what I am paid to do." }, { "from": "human", "value": "Yeah." }, { "from": "gpt", "value": "I mean, it's, it's, you know, it's, it's great. And, um, I've been, I've been fortunate in so many ways. And one way in which I've been fortunate is, uh, my parents, my parents, um, Rachel Spann Kennedy, Henry Harold Kennedy, my mother born in Columbia, South Carolina. My father from, uh, uh, New Orleans, Louisiana, they were refugees from the Jim Crow South. Uh, they were people who put their all into their children. Um, I have an older brother of a younger sister. All three of us know beyond, you know, beyond any controversy that we were loved and dearly loved by our parents and they were great people. My father, um, very interesting man, very independent minded. He was perfectly willing to go his own way. And, uh, I learned much from him, including I learned, I learned things from him, even when I ultimately disagreed with him. So for instance, again, to go back to his pessimism. Yeah, he was pessimistic, thoroughly pessimistic. Um, but he was also, he was also willing to change in certain ways. In fact, both my parents, I mean, one of the most important things that happened to my parents was that I'd say when I was, let's say 10 years old, I was born in 1954. So in 1964, in 1964, I think my parents would have taken the position that, um, you definitely never under any circumstances trust a white person. If a black person trusts a white person, that person is a fool. Do not trust white people. All right. They are not to be trusted. And, um, it's the highly, highly, highly unusual one who is not prejudiced. Okay. So white people are not to be trusted and, you know, by and large are going to be your enemy. All right. So let's just face that. That's, that's was their, their point of view. I'd say that, I'd say that 10 years later, that point of view had been leavened somewhat. Um, I think, and you know, you say, well, what happened in those 10 years? Well, certainly in my life, uh, one of the things that happen in those 10 years is I, I was a student in various schools and I had a series of teachers. I've had wonderful black teachers. I've also had wonderful white teachers and my parents paid attention to who my teachers were and how my teachers treated me. And I think that they were affected by the way in which there were white people who really helped me and were on my side and were thoroughly on my side. And I think that that experience, uh, changed my parents in their general view. They, you know, skeptical. Yeah, but they were, um, the, the, the, the possibility, the possibility of a white person genuinely being the friend of a black person that they, they, they, that became alive to them and I give them a lot of credit because they were, they were adults and, you know, for an adult to change, that's a big deal, but" } ]

[ { "from": "human", "value": "my, my parents did change in that transformation that was inspiring to" }, { "from": "gpt", "value": "you was formative to you in terms of joining the optimist camp, hugely. And, you know, uh, the, the, the school, I mean, I, I, again, you know, I said a moment ago, how lucky I've been, heck, um, I teach it, I teach at Harvard law school, I attended, I ended Balliol college, Oxford. I got my law degree from Yale law school. I got my undergraduate degree at Princeton university. Those are some pretty good schools. They all were, they all were the most important school. However, that I attended the school that made the most difference in my education was my high school, St. Albans school for boys and at St. Albans I encountered a cadre of teachers. And by the way, uh, St. Albans when I went to St. Albans all these teachers were white, all of these teachers. There was one, the, the, the, the, the head of athletics, very important man, very impressive man, Brooks Johnson, was black. Otherwise white teachers. And these teachers made a huge difference in my life. And, you know, I can call their names, boom, boom, boom, boom, boom. The greatest of them all was a man by the name of, um, Jack McKeown. Yeah. We called him gentlemen, Jack, gentlemen, Jack McKeown. And, um, what subject did he teach? He taught me, uh, history. He was my advanced placement history teacher, but, um, John F. McKeown and we shared a birthday. Um, John F. McKeown was a fabulously good teacher and we developed a deep lifetime friendship. I was with Mr. McKeown the day before he died and, um, you know, he was a white man and, um, I've had other teachers who've some of whom have become colleagues of mine. Uh, you know, Sanford Levinson Sanford Levinson was a teacher of mine at Princeton. Uh, he's become a colleague of mine. I mean, it would be, frankly, it would be impossible for me to, you know, I hear P I can't make some sort of, you know, sort of blanket condemnation of white people with Sandy Levinson in my life. Well, John F. McKeown in my life with Eric Foner in my life with my colleague Martha Minow in my life, my colleague Cass Sunstein in my life, impossible." }, { "from": "human", "value": "You know, that speaks to the power of teachers and mentors." }, { "from": "gpt", "value": "Oh yeah." }, { "from": "human", "value": "And you're that to a lot of people." }, { "from": "gpt", "value": "Well, I, I've, I've taken that to, I hope, I hope, listen, I would be, I would listen, I would be absolutely overjoyed if there was a student who thought of me in the way that I think of gentlemen, Jack McKeown, if there, if, if, if that's the case, eat just one, just one, if that's the case, I'm overjoyed." }, { "from": "human", "value": "That's a, that's a life well-lived. What do you think, uh, of Martin Luther King's, uh, I have a dream." }, { "from": "gpt", "value": "You still share that dream? I think that Martin Luther King juniors, I have a dream speech is one of the great speeches, not only in American history, but in the history of the world. You know, there's, there's a tendency now for people to sort of poo poo that speech. Um, I think unfortunately the speech, you know, has been embraced by, you know, you know, advertisers and corporate America. It's been, it's been so, you know, it's, it's been, it's been heard so many times that it, it's sort of has been made to suffer from what some people might view as overexposure." }, { "from": "human", "value": "It's so unfortunate. I mean, uh, it's too bad. Well, when, uh, especially when Hollywood rolls in, they did that with, uh, John Lennon's Imagine, uh, recently, uh, which I think is one of the greatest songs ever, and the actors and actresses ruined it by trying to like sing along" }, { "from": "gpt", "value": "and do this kind of cliche Hollywood thing, uh, recently, uh, recently thing. It has, but people have tried to make it into a cliche. The fact of the matter is it's the, the, the sentiment, the sentiment behind, I have a dream. Yeah, I'll associate myself with that. And anybody wants to, anybody wants to see some great oratory go watch a watch. And, um, and, um, you know, Martin Luther King Jr. I mean, he, he gave, he gave, you know, several great speeches, you know, his first speech. The first speech that he gave as a civil rights leader, the Holt street Baptist church, 1955, at the beginning of the Montgomery bus boycott, which was virtually an ex-temporaneous speech was one of his greatest. And he was, he was, I think 26 years old, great. His last speech, you know, his mountaintop speech, great speech, but I have a dream. I love it. Um, I will associate myself with those sentiments any day of the week." }, { "from": "human", "value": "Absolutely. So you still have hope for, um, a deep kind of, uh, multiracial," }, { "from": "gpt", "value": "a deep unity in the 24th century. I have that hope. And I think that the, I think the sentiments that Martin Luther King Jr. expressed in August, 1963, that represents the best of American life. And, um, I think it's a, you know, do I want to see that come to pass? Yeah. I want to see that come to pass and we'll, we'll work to, you know, push" }, { "from": "human", "value": "that project along as far as it'll go. Well, thank you for carrying his spirit of optimism forward, uh, the spirit of your mother and father. Uh, thank you for all the amazing work you've done and thank you for just this conversation is a huge honor. This is awesome. I really." }, { "from": "gpt", "value": "Thank you, Randall." } ]

[ { "from": "human", "value": "Thank you. Thanks for listening to this conversation with Randall Kennedy to support this podcast, please check out our sponsors in the description. And now let me leave you with some words from Martin Luther King Jr. Darkness cannot drive out darkness. Only light can do that. Hate cannot drive out hate. Only love can do that. Thank you for listening and hope to see you next time." } ]

[ { "from": "human", "value": "To support it, please check out our sponsors in the description. And now, dear friends, here's Anna Verbel. Let's go back to the early days. What did the formation of the Milky Way galaxy look like? Or maybe we want to start even before that. What did" }, { "from": "gpt", "value": "the formation of the universe look like? Well, we scientists believe there was the Big Bang, some big beginning. But what is important for my work and I think that's what we're going to talk about is what kind of elements were present at that time. So the Big Bang left a universe behind that was made of just hydrogen and helium and tiny little sprinkles of lithium. And that was pretty much it. As it turns out, it's actually quite hard to make stars or any structure from that. That's fairly hot gas. And so the very first stars that formed prior to to any galaxies were very massive stars, big stars, 100 times the mass of the Sun. And they were made from just hydrogen and helium. So big stars explode pretty fast after a few million years only. That's very short on cosmic timescales. And in their explosions, they provided the first heavier elements to the universe because in their cores, all stars fuse lighter elements like hydrogen and helium into heavier ones. And then that goes all the way up to iron. And then all that material gets ejected in these massive supernova explosions. And that marked a really, really important transition in the universe because after that first explosion, it was no longer chemically pristine. And that set the stage for everything else to happen, including us here talking today." }, { "from": "human", "value": "So what do you mean by pristine? So there's a whole complex soup of elements now as" }, { "from": "gpt", "value": "opposed to just hydrogen, helium, and a little bit of lithium of elements. Yeah. So after the Big Bang, just hydrogen and helium, we don't really need to talk too much about lithium because the amount was so small. And after these very first stars formed and exploded, they, the heavier elements like carbon, oxygen, magnesium, iron, all of that stuff was suddenly present in the gas clouds. Tiny amounts only, very tiny amounts. But that actually helped, especially the carbon and the oxygen, to make the gas cool. These atoms are more complicated than hydrogen. That's just a proton. And so it has cooling properties, can send out photons outside of the gas cloud so the gas can cool. And when you have gas that gets colder and colder, you can make smaller and smaller stars. So you can fragment it and clump it and turn it into stars like the Sun. And the cool thing about that is that when you have small stars like the Sun, they have a really long lifetime. So those first low mass stars that formed back then are still observable today. That is actually what I do. I try to find these early survivors because they tell us what the gas looked like back then. They have preserved that composition of these early gas clouds, the chemical compositions, until today. So I don't need to look very far into the universe to study all the beginnings. I can just chemically analyze the oldest stars. And it's like unpacking everything" }, { "from": "human", "value": "that happened back then. It's very exciting. So to just reiterate, so in the very early days in the first few million years, there was giant stars that's mostly hydrogen and helium, and then they exploded in these supernova explosions." }, { "from": "gpt", "value": "And then they made these clumps. Yeah, so the first stars... Not pristine clumps. Yeah, so the first stars... Yeah, pretty much fun. So it took a few hundred million years for the first stars to emerge, and then they exploded after a few million years. Kaboom. And then it's like I always consider the universe like a nice soup, and then these first supernova explosions kind of provided the salt, you know, just a little sprinkle of heavier elements, and that made it really tasty. It just changed it completely, right? And that changed the physics of the gas. So that meant that these gas clouds that were, you know, surrounding the former first stars, they could now cool down and clump and form the next generation of stars that now included also little stars. And as I just mentioned, the small stars have these really long lifetimes. The Sun has a lifetime of 10 billion years. Any star that is even less massive will have an even longer lifetime. So that gives us a chance to still observe some of the stars that formed back then. So we are testing the conditions, the chemical and physical conditions of the early universe even before the galaxy formed." }, { "from": "human", "value": "Fun. So what's the timeline that we're talking about? What is the age of the universe and what is the earliest time we got those salty delicious soup, clump" }, { "from": "gpt", "value": "soups with heavier elements? Well, the universe is 13.8 billion years old. Allegedly, yeah. Well, when I was in high school, the universe was 20 billion years old." }, { "from": "human", "value": "So the estimate, do you think that estimate will evolve in interesting" }, { "from": "gpt", "value": "ways or no? I think we have mostly converged, yes, because the techniques are very different now, much more precise. The whole business of cosmology by mapping out the cosmic microwave background. That's a marvelous feat. Maybe the digits will still move around a little bit, but that's all right." }, { "from": "human", "value": "The gravitational waves and all that's all the different sources of data kind of" }, { "from": "gpt", "value": "mapping out this detailed picture of the early universe, that's all this mapping out this detailed picture the early universe. Totally. And so, we think the earliest little stars formed, I don't know, maybe half a billion years after the Big Bang, right? Again, a few hundred million years for the first stars to emerge and then, you know, took some time. So, give or take half a billion years. And that was the time when sort of the very first proto galaxies formed, early stellar structures, stellar systems from which the Mickey way eventually formed, right? So, it was the Mickey was probably a bigger, slightly bigger one. And we know today that galaxies grow hierarchically, which means they eat their smaller neighbors. So, if you're the bigger one and have a few friends around, you're just going to eat them, absorb them, and then you grow bigger. And so, all these little early stars, you know, kind of came into the Mickey way through that kind of process. And that's why we find them in the outer parts of the galaxy today, because they're just kind of deaf and just left there since." } ]

[ { "from": "human", "value": "So, the old stuff is on the outskirts of the galaxy and the new stuff is in closer to the middle? Is there? Broadly speaking, yes. So, that's where you would look for it." }, { "from": "gpt", "value": "So, maybe just to step back, like, what is a galaxy? What is a proto galaxy? Is there? Broadly speaking, yes. I love that question. So, the galaxy is a huge assembly of stars. The Mickey way contains something like 200 to 400 billion stars. And most of the material and the stars are in the disc. And when we look at the night sky, what we see as the Mickey way band on the sky, that is actually the inner, the next inner spiral arm, because we actually live in a spiral disc galaxy, so the Mickey way is a spiral disc galaxy. And we are looking, actually, depends a little bit in the Northern Hemisphere, we're looking out of the galaxy. So, we're seeing the next outer spiral arm. And as you can imagine, there's only dark space behind that. So, we don't see it all that nice on the sky. But if you travel to the Southern Hemisphere, let's say, South America, you see the Mickey way and it looks so different on the sky, because that's the next inner spiral arm, and that's backlit by the galactic center. The galactic center is a very big, puffy region of gas. There's a lot of star formation that the galactic party is happening there. So, it's very bright and it makes for this very beautiful Mickey way on the night sky that we see. So, actually, if you ever get the chance to experience that, I encourage you to almost close your eyes while seeing this and imagining that you're sitting in this kind of disc, in this pancake, and you're just kind of looking right into it. And you can really feel that we're in this 2D disc. And then you can imagine that there's a top and a bottom, and that we're really part of the galaxy. You can really experience that. We're not just lost in space somewhere, but we're really a part of it. And knowing a little bit about the structure of" }, { "from": "human", "value": "the Mickey way really helps. Do you feel small when you think about that? When you look on the spiral on the inside of the Milky Way, and then you look out to the outside? How are we supposed" }, { "from": "gpt", "value": "to feel the spiral? I don't know. I don't feel small necessarily. I feel in awe and I feel I'm a part of it because I can really feel that I'm a part of it. I think for many people, they think like, oh, there's just the planet and then there's nothing. And that's almost a little bit sad, but that's really not the case, right? Because there's so much more and I really like to imagine, wow, I'm sitting in this big galactic merry-go-round and we're going around the center and I can see the center above me, right? And I can almost feel like we're going there." }, { "from": "human", "value": "Of course, we can't really feel that, but the sun does circle the galactic center. But there's a kind of sadness to like looking pictures of a nice vacation place. All we get is that light, an old light. Do you feel sad that we don't get to travel or you and I will not get to travel there and maybe humans will never get to travel there?" }, { "from": "gpt", "value": "Yeah, I always wanted to travel into space and see the Earth and other things from up there. They're certainly bad, but I don't know. It's also okay. It would just be at our vantage point" }, { "from": "human", "value": "and see it from here. With the senses, with the telescopes that we have and explore the possibility. I mean, there is a kind of wonder to the mystery of it all. What's out there? What interesting things that we can't possibly imagine. There could be all kinds of life forms, bacteria, all this kind of stuff. I tend to believe that it depends on the day. I tend to believe there's just a lot of very primitive organisms just spread out throughout and they build their little things like bacteria type organisms. Just to think what kind of worlds there are because they're probably really creative living organisms. I guess the question I'm wondering to myself when I look out there to the stars, how different are the conditions on the different" }, { "from": "gpt", "value": "planets that orbit those stars? It will definitely be very different. I mean, the variety out there is huge. We know now that I think it's about every other star has at least one planet. I already mentioned the number of stars in the galaxy. I mean, it's a huge number of planets out there. Who knows what that looks like? All we know is that there is a lot of variety. We don't quite yet understand what drives that, what governs that," }, { "from": "human", "value": "why that is the case, why is it not all one size fits all. I mean, the dynamics of planet" }, { "from": "gpt", "value": "formation, like exoplanet formation or star formation, the whole. All of it. Star formation remains a much research topic. We definitely know that it works because all the stars are there. Same for the planets, but the details are so varied per gas cloud. It's very hard to come up with very detailed prescriptions. Broadly, we have figured it out. You need a gas cloud. You need to cool it, something clumps and fragments, and somehow it makes a star with planets or without. But the dynamics of the clumping process is not fully understood. No, no. The local conditions are so varied. I mean, it's the same with all people look like" }, { "from": "human", "value": "people, but individually we look very different. Even the subtle diversity of the formation process" }, { "from": "gpt", "value": "creates all kinds of fun differences. We just don't know how this turned out in an individual case. It's kind of hard to figure it all out and to take a look, certainly with planets, the chance to ever actually take a picture of a planet. It's miniscule because they don't shine, so they're really dark. I'd say there's a lot of possibility out there," } ]

[ { "from": "human", "value": "but we have to be a little bit more patient before we come up with technologies where patience becomes less necessary by extending our lifetimes or increasing the speed of space travel, all the kind of stuff. Humans are pretty intelligent. They're pretty intelligent. For the most part, I hope, when I'm on the optimistic days. Well, maybe just to linger on what a galaxy is. What should we know about our understanding of black holes in the formation? Is that an important thing to understand in the formation of a galaxy? So all the orbiting, all the spiraling" }, { "from": "gpt", "value": "that's going on, how important is that to understand? All of the above. That's what makes astronomy really hard but also really interesting. No day is like another because we always find something new. I want to come back to the idea of the proto-galaxy because it actually matches or relates to the black hole formation. So most large, well, pretty much all large galaxies have a supermassive black hole in the center. We don't really know where they come from. Again, we know that they are there but how do we get there? So we go back to the early universe. We had a little galaxy that just sort of, I don't know, had some small number of stars. It was a first gravitationally bound structure that was held together by dark matter because dark matter actually kind of structured up first before the luminous matter could because that's what dark matter kind of does. And it started to hold gas and then stars sort of together in these first very shallow, what we call potential wells, so these gravitationally bound systems. And then the Milky Way grew from absorbing neighboring smaller, even smaller systems. And somewhere in that process, actually, there must have been a seed for one of these supermassive black holes. And I'm not actually sure that it's clear right now kind of what was there first, the supermassive black hole or the galaxy. So lots of people are trying to study that. And of course, the black hole wasn't as massive back then as it is these days. But it's a big area of research and the new James Webb, the JAWST, the telescope, the infrared telescope in space is working on, many people are working on that to figure out exactly what happened. And there are some surprising results that we really don't understand right now." }, { "from": "human", "value": "So to solve the chicken or the egg problem of, do you need the supermassive black hole to form a galaxy or does the galaxy naturally create the supermassive black hole?" }, { "from": "gpt", "value": "Yeah, I mean, I think to some degree, we can answer that because there are lots of little dwarf galaxies out there. The Milky Way remains surrounded by many dozens of small dwarf galaxies. I have studied a bunch of them. And to the extent that we can tell, they do not contain black holes. So they certainly were gravitationally bound structures. So either you can call them proto galaxies or dwarf galaxies or first galaxies, they were definitely there. But they must have been bigger things like the proto Milky Way, where something was different. What made them more massive so that they would gravitationally attract these smaller systems to integrate them." }, { "from": "human", "value": "So we'll have to see. How do we look into the dynamics of the formation, the evolution of the proto galaxies? Is it possible? Did they shine? I mean, what are the set of data that we can possibly look at? So we've got gravitational waves, which is really insane that we can even" }, { "from": "gpt", "value": "detect those. There's light. What else can we detect? So that would fall into the category of observational cosmology. And the JWST is the prime telescope right now. And it promises big, big steps forward. This is in its early days, because it's only been online like a year or so. But that collects the infrared light from the farthest, literally proto galaxies, earliest galaxies. That light has traveled some 13 billion years to us. And they're observing these faint little blobs. And folks are trying to, again, study the onset of these early supermassive black holes, how they shape galaxies. So they're seeing that they were there, surrounded by already bigger galaxies. Ideally, I'd like for my colleagues to push a little bit further. Hopefully, that will eventually happen. In terms of looking towards the older and older ones. Yeah. More and more primitive in terms of the structure. But of course, as you can imagine, if you make your system smaller and smaller, it becomes dimmer. And it's further and further away. So we're reaching the end of the line from a technical perspective pretty quickly." }, { "from": "human", "value": "But dimmer and dimmer means older and older." }, { "from": "gpt", "value": "Yes, in a sense, because it all started really small. Because it's small or smaller, which correlates to older and older. In that phase of the universe, it would, otherwise, it doesn't." }, { "from": "human", "value": "Yeah, because it is small or smaller, which correlates to older and older. Just to take a small attention about black holes, you know, because you do quite a bit of observational cosmology and maybe experimental astrophysics. What's the difference to you between theoretical physics and experimental? So there's a lot of really interesting explorations about paradoxes around black holes and all this kind of stuff. About black holes destroying information. Do those worlds intermix to you?" }, { "from": "gpt", "value": "Especially when you step away from your work and kind of think about the mystery of it all. Well, at first glance, there isn't actually much crosstalk. Personally, I mostly observe stars. So I don't usually actually think too much about black holes." }, { "from": "human", "value": "And stars is a fundamentally kind of chemical, physical phenomena that doesn't..." }, { "from": "gpt", "value": "That's right. The physics is kind of different. It's not extreme. I mean, you know, you could consider nuclear fusion sort of be perhaps extreme. You need to tunnel. That has some interesting physics there. But it's just a different flavor. I don't do these kinds of calculations myself either. I very much like to talk with my theory colleagues about these things, though, because I find there's always an interesting intersection. And often it's just... I've written a number of papers with colleagues who do simulations about galaxies. And so they're not quite as far removed as, let's say, the black hole pen and paper folks. But even in those cases, we have the same interests in the same topics, but it was almost like we're speaking two different languages. And we weren't even that far removed, both astronomers and all. And it was really interesting just to take the time and really try to talk to each other. And it's amazing how it's fair, but it's as how hard that is. You know, even amongst scientists, we already have trouble talking to each other. Imagine how hard it is to talk to non-scientists and other people to try to... We're all interested in the same things as humans at the end of the day, right? But everyone has sort of a different angle about it and different questions and way of formulating things. And sometimes it really takes a while to converge and to get to the common ground. But if you take the time, it's so interesting to participate in that process. And it feels so good in the end to say like, yes, we tackled this together, right? We overcame our differences, not so much in opinion, but just in expressing ourselves about this and how we go about solving a problem. And these were some of my most successful papers, and I certainly enjoyed them the most." } ]

[ { "from": "human", "value": "It could also lead to big discoveries. You put it really well in saying that we're all kind of studying the same kind of mysteries and problems. I mean, I see this in the space of artificial intelligence. You have a community, maybe it seems very far away, artificial intelligence and neuroscience. You would think that they're studying very different things, but one is trying to engineer intelligence, and in so doing, try to understand intelligence, and the other is trying to understand intelligence and cognition in the human mind. And they're just doing it from a different set of data, a different set of backgrounds, and the researchers that do that kind of work. And probably the same is true in observational cosmology and simulation. So it's like a fundamentally different approach to understanding the universe. Let me use, for simulation, let me use the things I know to create a bunch of parameters and create some, just play with it. Play with the universe. Play God. Create a bunch of universes and see in a way that matches experimental data. That's a fun, it's like playing Sims, but at the cosmic level. Yes, yes. And then probably the set of terminology used there is very different, and maybe you're allowed to break the rules a little bit more. Let's have, yeah, take the Drake equation. Yeah, you don't really know. You kind of come up with a bunch of values here and there, and just see how it evolves. And from that kind of intuit the different possibilities, the dynamics of the evolution of a galaxy, for example. Yeah, but it's cool to play between those two, because it seems like we understand so little about our cosmos." }, { "from": "gpt", "value": "So it's good to play and create a good level. Yes, it's like a big sandbox, right? And everyone kind of has their little corner, they do things, but we're all in the same sandbox together at the end of the day." }, { "from": "human", "value": "But in that sandbox does have super powerful and super expensive telescopes. That everybody's also, all the children are fighting for the resources to make sure they get to ask the right questions using that big cool tool. Well, can we actually step back on the big field of stellar archaeology? What is this process? Can you just speak to it again?" }, { "from": "gpt", "value": "You've been speaking to it, but what is this process of archaeology in the cosmos? Yeah, it's really fascinating. So I mentioned the lesser the mass of the star, the longer it lives. And again, for reference, for the next dinner party, the sun's lifetime is 10 billion years. So if you have a star that's 0.6 or 0.8 solar masses, then its lifetime is going to be 15 to 20 billion years. And that's an important range for our conversation because again, if you assume that such a small star formed soon after the Big Bang, then it is still observable today. You mentioned old light before. Yeah, that light is like a few thousand years old, but compared to the age of these stars, it's nothing. So to me, that's young. It comes straight from our galaxy. These stars are not far away. They're now a galaxy in the outskirts. They probably did not form in the galaxy because again, a hierarchical assembly of a Milky Way meant... Exactly. They formed in a little other galaxy in the vicinity. And at some point, the Milky Way ate that, which means it absorbed all the stars, including these little old stars that are now in the outskirts of the Milky Way that I used to point my telescope to. So Milky Way meant what can we learn from these stars? Why should we study them? Now, these little stars are really, really efficient with their energy consumption. They're still burning for the experts, just burning hydrogen to helium in their cores. And they have done so for the past 12, 13 billion years, however old they are, and they're going to keep doing that for another few billion years, same as the sun. The same sun also just does hydrogen to helium burning and will continue that for a while, which means the outer parts of the star, well, pretty much actually most of the star, that gas doesn't talk to the core. So whatever composition that star has in its outer layers is exactly the same as the gas composition from which the star formed. Which means it has perfectly preserved that information from way back then all the way to today and going forward. So I'm a stellar archaeologist because I don't dig in the dirt to find remnants of past civilizations and whatnot. I dig for the star, for the old stars in the sky because they have preserved that information from those first billion years in their outer stellar atmosphere, which is what I'm observing with telescopes. So I'm getting the best look at the chemical composition early on that you could possibly wish for." }, { "from": "human", "value": "What kind of age are we talking about here? Are we talking about something that's close to that," }, { "from": "gpt", "value": "like a 13 billion, 12, 13 billion age range here, age range? That's what we think. Now, there's a small caveat here. We cannot accurately date these stars, but we use a trick to say, oh, these stars must have formed as some of the earliest generations of stars because we need to talk about the chemical evolution of the universe in the Milky Way for a second. So I already mentioned the pristineness of the universe after the Big Bang, right? Just hydrogen and helium. Then the first stars formed, they produced a sprinkle of heavier elements up to iron. Then the next generation of stars formed. That included, again, massive stars that they would explode again, but also the little ones that keep on living. And then the massive ones, again, explode as supernovas, so they provide, again, another sprinkle of heavier elements. And so over time, all the elements in the periodic table have been built up. There have been other processes, for example, neutron star mergers and other exotic supernovae that have provided elements heavier than iron all the way up to uranium from very early on. We're still trying to figure out those details, but I always say pretty much all the elements were done from day three." } ]

[ { "from": "human", "value": "So iron is where, once you get to iron, you got all the fun you need." }, { "from": "gpt", "value": "Most of the fun you need. Most of the fun. Yes and no. I really like the heavier elements. Gold, silver, platinum, that kind of stuff." }, { "from": "human", "value": "For personal reasons or for star formation?" }, { "from": "gpt", "value": "Well, both." }, { "from": "human", "value": "What's the importance of these heavier metals in the evolution of the stars?" }, { "from": "gpt", "value": "So they're the spice of life, right? So every supernova gives you elements up to iron. That's cool, but at some point, it gets a little bit boring because that always works. But that's the baseline. We need that. That's certainly what came out of the first stars and then all the other supernova explosions that followed with every generation. It took about 1,000 generations, give or take, until the sun was made. So the sun formed from a gas cloud that was enriched by roughly 1,000 generations of supernova explosions. That's why the sun has the chemical composition that it has, including, you know, and somehow the planets were made from that as well." }, { "from": "human", "value": "So the supernova explosions, the many generations," }, { "from": "gpt", "value": "are creating more and more complex elements. No, it just goes all the way up to iron. And then it's a little bit more of all of these elements. Just more. Yeah, it's one sprinkle, then another, and it just kind of adds up, right? Now, the heavy elements form in very different ways. They're not fusion-made. They are made typically through neutron capture processes, but for that, you need seed nuclei, ideally, you know, iron or carbon or something. So the supernova-made elements are very good seed nuclei for other processes that then create heavy elements. And because they cannot be made everywhere, they, when you, when, you know, so I, some of my stars have huge amounts of these heavy elements in them. And they tell us in much more detail, something really interesting happened somewhere." }, { "from": "human", "value": "Just more. Well, wait, I thought the really old ones we would not have. So what does that mean if the old?" }, { "from": "gpt", "value": "Yes, important clarification. So the stars that we are observing today, these old ones, they formed from the gas. And the question is, what enriched that gas? So it could have been just a first star dumping" }, { "from": "human", "value": "their elements into that gas all the way up to iron." }, { "from": "gpt", "value": "And we have found some stars that we think are second generation stars. So they form from gas enriched by just one first star. That's super cool. Then we find other old stars that have a much more complicated heavy element signature. And that means, okay, they're probably formed in a gas cloud that had a few things going on, such as maybe a first star, maybe another more normal supernova, and maybe some kind of special process, like a neutron star merger that would make heavy elements. And so they created a local chemical signature from which the next generation star then formed. And that is what we're observing today. So all these old stars basically carry the signature from all these progenitor events. And it's our job then to unravel, okay, which processes and which events and how many may have occurred in the early universe that led to exactly that signature that we observed 13 billion years later." }, { "from": "human", "value": "Is it possible to figure out the number of generations that resulted in these stars?" }, { "from": "gpt", "value": "Well, we think we can sort of say, okay, this was like second generation or third, because the amounts of heavy elements in the stars that we observe is so tiny. One super, one normal supernova explosion is actually already basically too much. It would give us too much of it. And the thing is, you can never take away things in the universe. You can only add, there's no cosmic vacuum cleaner going around sucking things away. Black holes are probably the closest to that, but they would have taken the whole star. Yeah, they would take the whole thing not just a little bit. They would take stuff out of the gas, you know. So we have maybe 10 stars or so now where we are saying that they contain so little of these heavy elements that there must be second generation because..." }, { "from": "human", "value": "Yeah, they would take the whole thing not just a little bit." }, { "from": "gpt", "value": "How else would you have made them? And again, I want to stress that the elements that we observe in these stars were not made by the stars themselves that we observe. That's just a reflection of the gas cloud. So we aren't actually, I had to say, I love stars. At the end of the day, we don't really care for the stars that we're observing. We care for the story that they're telling us about the early universe." }, { "from": "human", "value": "So, yeah. So the stars a small mirror into the early universe." }, { "from": "gpt", "value": "Yeah." }, { "from": "human", "value": "And so what are you detecting about those stars? Can you tell me about the process of archaeology here? Like what kind of data can we possibly get to tell the story about these heavy elements on the stars?" }, { "from": "gpt", "value": "Yeah, it depends really on what star you find. There are many different chemical signatures. We actually pair up these days our element signatures with also kinematic information, how the star moves about the galaxy. That actually gives us clues as to where the star might have come from. Because again, all these old stars are in the galaxy, but they are not off the galaxy. That's a small but important distinction. So they all came from somewhere else." }, { "from": "human", "value": "So you can rewind back in time," }, { "from": "gpt", "value": "kind of estimate where it came from. Yeah. So we kind of want to say, oh, it came from that in that dwarf galaxy. But interestingly enough, so I'm just a few days ago, I submitted a paper with three women undergrads. It was so good to work together. And we found a sample of stars that have very, very low abundances in strontium and barium. So very heavy elements. And I had a hunch for a while that these stars would probably be some of the oldest. Because, as I said, heavy elements give you extra information about special events. And again, finding something that's really low means that must have happened either really early on or in a very special environment. Because we can only ever add. So if you find something that's incredibly low in terms of the abundance, it maybe just one event contributed that, max. So we looked at the kinematics. How are these stars moving? And they're all going the wrong way in the galaxy. How is that possible? Well, it is possible because consider, now we come back to the proto-galaxy. The proto-galaxy was like a beehive. It just didn't really know what it was or what it wanted to become when it grew up. And it was absorbing all these little galaxies to grow fast. Some galaxies, some absorbed galaxies were thrown in going the main way. And some came in the wrong way. Happens. Yeah, it happens. But this could only happen early on when there wasn't left and right and up and down. So stuff would come in from all ways. So now, 13 billion years later, we're looking- They're still doing it. Yeah, they're still doing it. And B, we just looked for stars that have low strontium and barium abundances. And then we look at the kinematics. And lo and behold, they're all at hundreds of kilometers per second going the wrong way. It's like, dude, you must have come in really early on from somewhere else. So we call this retrograde motion. That's a clear sign of accretion. So something that has come in to the galaxy. And because they are so fast, and it's really all of them, that must have happened early on. You can't throw a galaxy into the Mickey Ray right now the wrong way. It eventually will turn around." } ]

[ { "from": "human", "value": "We're still doing it. Can you actually just, in a small tangent, speak to the three women undergrads, like this little- It's pretty cool that you were able to use a hunch to find this really cool little star. Yeah, what's the process like? Especially with undergrads, I think they'll be very interesting and inspiring to people." }, { "from": "gpt", "value": "Yes, it was a wonderful little collaboration that actually emerged in the fall. I really like working with undergrads and grad students, postdocs. And I came up with a new concept for a class at MIT where I wanted to integrate the research process into the classroom. Because sometimes people find it really hard to call email a professor, I'm this and that person, and I'm interested in your research. Could I possibly come? I wanted to streamline that and give, and not just trial, how it would work to provide the safe confines of a classroom where you just sign up and do research in a very structured way. I developed it. It was a lot of work, a little bit more than I thought, to map up an entire research project basically from scratch in 10 worksheets so that they could do it, again, in a very structured and organized fashion. Also, you created this whole framework for it for them to do it. Yes, the whole thing. But the promise was, you come sign up for my class, in Teams of Two, you each get your own old star that has not been analyzed before. I don't know what the solution is, because in research, we don't look up the solution at the end of the book. We do not know what we're going to find. Our job is to do the work and then to interpret the numbers, because our job as scientists is to find the story. Anyone can crunch numbers. Anyone." }, { "from": "human", "value": "Also, you created this whole framework" }, { "from": "gpt", "value": "for it for them to do it. It's complicated sometimes, but it's doable, right? But coming up with a story, when you only have three puzzle pieces, what does the puzzle look like?" }, { "from": "human", "value": "It's complicated sometimes," }, { "from": "gpt", "value": "but it's doable. You have to be a little bit bold. You need to have some experience, and you need to kind of see the universe in 3D. You just need to kind of go for it. And that's the beautiful thing. I really love that." }, { "from": "human", "value": "And so this was a story of weird kinematics going the wrong way, combined with this particular weird signature in terms of the heavy elements. Exactly. And you have to come up with a story" }, { "from": "gpt", "value": "about those heavy elements. Exactly. And you have to come up with that. And so the story of that paper is now, usually I don't say I find the oldest stars. When I talk to my research colleagues, I talk to them about we find the chemically most pristine stars, because that's actually what we measure, the chemical abundance that tells us, okay, it must have been second or third or fifth generation of stars, right? But these low strontium stars that go in the wrong way, like they're getting paid for it, they must be the oldest stars that came into the galaxy because they formed before the galaxy was the Mickey way, right? And it's so cool. And it was so wonderful. So this class, it went so well in the fall. I had nine people sign up. That's not unusual for a class at the specialty class at MIT, so small number. It was eight women and they were so into it that I said, okay, let's use this opportunity. You're going to do some extra work with me and we're going to publish this. Try to publish." }, { "from": "human", "value": "Try to publish. Yes. I also like that you're using the terminology of chemically more pristine. When I'm talking to younger people, I'll just say that I'm more chemically pristine than them. I like that description of age. So there's this term of metal, poor stars. So most of these old stars are going to be metal poor." }, { "from": "gpt", "value": "Yes. I searched for the most metal poor stars." }, { "from": "human", "value": "And what is that? Can we just define? Yeah, what does that mean?" }, { "from": "gpt", "value": "Yeah, what does that mean? I don't know who came up with this. I would love to know, but the universe is a complicated place. So many decades ago, someone clever came up with the idea to say, let's simplify things a little bit. Let's call hydrogen X, helium Y, and all the other elements combined, metals, Z. Oh, when I give public talks, I always ask, is there a chemist in the audience? Let me just tell you, neon is a wonderful metal. And they're like, oh my God, what's he saying? I'm an astronomer. I'm not a chemist, so I'll get away with it. So if you just roll with it for a moment, all the elements except hydrogen and helium are called metals. Now, if we look again at chemical or the concept of chemical evolution, it means more and more of all the elements, everything heavier than hydrogen and helium gets produced slowly but surely by different types of stores and events. So that's a monotonously increasing function. And so we look for the stars that have the least amounts of heavy elements in them, because that means we are going further and further back in this process, in that function, almost all the way to the very beginning. And that is the first source, right? They started that process. That's why I said it was such an important transition phase because we call the post-Big Bang universe pristine, just hydrogen and helium. And after that, the mess started. As soon as you add elements to it, things kind of get a little out of hand. That ends in this beautiful variety that we have everywhere these days." } ]