2009/01/02/#ubuntu-devel.txt

[00:00] <kolby_> There's not really any college classes to take for open source developers are there?
[00:07] <Laney> kolby_: Well it's more about getting stuck in than taking classes I suppose
[00:40] <directhex> Laney, IMHO that's the case with free software work generally
[00:40] <Laney> right, that's what I meant
[00:41] <directhex> Laney, i didn't get my cushy linux sysadmin job, with root on a couple of million quid of kit, on the basis of stuff i learnt in classes. i got it on the basis of the linux mucking about i did instead of revising for exams! :)
[00:42] <Laney> theory != practice
[00:42] <directhex> well, yes. and so much of computing is a mix of both
[00:42] <Laney> mhm
[00:42] <Laney> kolby_: Pick a project, get your hands dirty :)
[00:42] <directhex> purely practice and, well, you get the mess of people who answer linuxquestions or ubuntuforums posts with things involving --prefix=/
[00:43] <directhex> purely theory, and you go nowhere fast either
[00:44]  * Laney only knows theory ;)
[00:44] <Laney> silly academia
[00:44] <Laney> "When was the last time you ran one of your programs?" is a joke in our lab :S
[00:44] <kolby_> what classes do you take?
[00:44] <Laney> none - I'm doing a PhD ;)
[00:44] <kolby_> amazing.
[00:44] <directhex> pfft, bloody students
[00:45] <Laney> quite
[00:45] <directhex> bane of my sodding life, the lot of you
[00:45] <kolby_> I have about... 1/2 a bachelor
[00:45]  * Laney files a support request with directhex
[00:45]  * directhex disparages Laney on rt-comment@
[00:46] <Laney> DATA PROTECTION REQUEST
[00:46] <kolby_> is socket programming necessary for network programming?
[00:47] <kolby_> or is it going too deep for simple game programming?
[00:48] <directhex> Laney, certainWHOOPS, i got my SELECT and DROP the wrong way round in postgres
[00:48] <directhex> kolby_, what kind of game? and i think we've deviated into #ubuntu-motu territory here
[00:48] <Laney> it's OK, you have highly redundant backups... right?!?!?
[00:49] <directhex> Laney, they were redundant, so i binned them!
[00:49] <Laney> foiled again
[00:49] <kolby_> directhex: so what is this channel for?
[00:50] <Laney> I'd say it's something like Development of Ubuntu (not support, not app development on Ubuntu)
[00:50] <kolby_> okay
[00:50] <Laney> #ubuntu or a general programming channel
[00:50] <directhex> better fit than here, albeit not ideal either
[00:50] <directhex> this is the lair of scary people like jono. he can devour you in one bite!
[00:50] <kolby_> maybe ##c
[00:50] <kolby_> jono sounds like a cool enough guy.
[00:51] <directhex> yeah, sure. but you'll go there without me. i don't do c.
[00:51] <kolby_> I still haven't heard jono's metal.
[00:51] <Laney> #haskell is where all the rock stars go to code
=== kolby_ is now known as directoct
[00:51] <directoct> directhex: mwahahaha.
[00:52] <jono> directhex, shame on you :P
[00:52]  * directhex tosses a bacon cookie @ jono 
[00:52] <jono> :)
[00:53] <directhex> chocolatey AND meaty!
[00:53] <directoct> Laney: haskell != game programmer friendly
[00:53] <Laney> not necessarily
[00:53] <directhex> c != friendly ;)
[00:53] <directhex> all depends on what you're doing, and where your bottlenecks are, of course
[00:54] <Laney> I wrote sudoku in it!
[00:54] <directoct> directhex: that's about as unhealthy as these "deep fried chocolate covered twinkies" I saw at the state fair.
[00:55] <directhex> bacon! http://www2.apebox.org/wordpress/wp-content/gallery/00-single/normal_IMG_0551.JPG
[00:55] <directoct> sudoku... has the _potential_ to be resource intensive.
[00:56] <directhex> pro-grade games are written in multiple languages
=== directoct is now known as kolby_
[00:56] <directhex> usually
[00:56] <kolby_> pro-grade = something like wow?
[00:56] <directhex> graphics engine in something like c, and all the game logic in something easier like lua, python, or a custom language (like unrealscript)
[00:57] <kolby_> directhex: ohhhhh...
[00:57] <kolby_> I should boiler-plate my code.
[00:58] <kolby_> I suppose it has been a trend.
[00:58] <directhex> as an example, Sid Meier's Civilization 4 uses Python 2.4 for all the game logic
[00:58] <kolby_> directhex: you lie beautiful lies.
[00:59] <directhex> /media/disk/Program Files (x86)/Firaxis Games/Sid Meier's Civilization 4/python24.dll: MS-DOS executable PE  for MS Windows (DLL) (GUI) Intel 80386 32-bit
[01:01] <savvas> (Literal reply: take that!)
[01:01] <kolby_> I have to tell you directhex, that's great news.
[01:01] <savvas> :)
[01:02] <directhex> many games use some Free Software somewhere
[01:02] <kolby_> directhex: ...so where's the credit?
[01:02] <directhex> kolby_, erm... hiding!
[01:02] <directhex> kolby_, in the back of the manual, generally
[01:03] <kolby_> why wasn't I informed about my nice open source dlls?
[01:03] <savvas> the python license doesn't expect credit I think, right?
[01:03] <directhex> dunno. i don't do python
[01:03] <directhex> vorbis.dll is a common one to find - all unreal tournament games since 2003, and some others (e.g. grand theft auto san andreas) use vorbis
[01:04] <kolby_> well...  they credit their stupid middleware.
[01:04] <directhex> as well as sony's singstar karaoke games for playstation
[01:04] <kolby_> directhex: I'm sure neversoft used it for their games.
[01:05] <directhex> possibly
[01:06] <directhex> generally, game developers have obscene deadlines & budget limits. they just go for the easiest route - best balance of cost & time
[01:06] <kolby_> directhex: nice wordpress site btw.
[01:06] <directhex> sometimes they opt for vorbis, sometimes they pay for mp3. sometimes they use python or lua, sometimes they pay devs to write something in-house
[01:06] <kolby_> strange cookies...
[01:06] <directhex> a few devs are using mono now for console development
[01:06] <directhex> and iphone
[01:07] <kolby_> at least C# has mono developer's respect
=== krusaf|bnc is now known as krusaf
[01:08] <kolby_> C# always seemed like M$'s answer to Sun's Java.
[01:08] <directhex> to an extent, it is
[01:09] <directhex> they got into trouble for making their "own" modified java, back in the early days of XP, so they took their ball & started their own game
=== moquist_ is now known as moquist
[01:10] <kolby_> learning new launguages can be fun.  Python was fun to learn.
[01:11] <directhex> i disagree on principle with whitespace sensitivity
[01:11] <kolby_> I like that idea.
[01:12] <kolby_> don't think of it as forced style.  Think of it as time well spent without "; punctuation"
[01:13] <kolby_> plus the code does look pretty.
[01:13] <directhex> hmm, i find python ugly
[01:13] <directhex> i like my {}!
[01:13] <kolby_> then again, I use underscores and full words in variabl names.
=== fta_ is now known as fta
[01:18] <kolby_> dont me this:  strange_function(crazy_array[x][y][z][r]);
[01:18] <kolby_> s/me/mind
[02:44] <bbechdol> Evening all
[02:46] <Laney> Can someone unsubscribe u-m-s from bug #310349, please?
[02:46] <ubottu> Launchpad bug 310349 in eog "eog does not handle nonadobe CMYK JPEGs correctly" [Undecided,New] https://launchpad.net/bugs/310349
[02:46] <Laney> erm, bug #313049 even
[02:46] <ubottu> Launchpad bug 313049 in smstools "Please merge smstools 3.1.3-1 from Debian experimental (main)" [Undecided,Confirmed] https://launchpad.net/bugs/313049
=== j_ack_ is now known as j_ack
[03:55] <ScottK-desktop> Laney: Done.
[04:23] <superm1> tjaalton, the "upstream installer" for fglrx builds the same debs that you would get from the archive.  (things that get uploaded to the archive are built with --buildpkg Ubuntu/source, while if you run it locally then --buildpkg Ubuntu/intrepid etc)
[04:53] <android60> are there plans to add support for atheros 5007eg cards in jaunty via the madwifi hal 10.5.6
[04:53] <android60> ?
[05:02] <atester> what packages do I need to install in order to get the binary nvidia drivers in Jaunty?
=== jonathan_ is now known as highvoltage
=== sepheebear is now known as Sepheebear
[08:02] <tjaalton> superm1: ok, you may reopen those if you like :)
=== asac_ is now known as asac
[13:19] <DktrKranz> StevenK, regarding bug 311706, TechBoard granted me upload privileges for SCons, is that sufficient to process the sync?
[13:19] <ubottu> Launchpad bug 311706 in scons "Please sync scons 1.2.0-1 (main) from Debian experimental (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/311706
[13:20] <StevenK> DktrKranz: I guess so. I'll add it to the processing list
[13:20] <DktrKranz> I don't know where uploads privileges are stored (no UI yet, IIRC)
[13:22] <DktrKranz> StevenK, if you have some spare time, mind NEWing amule-adunanza backport for intrepid and hardy (binary packages)?
=== warp10_ is now known as warp10
[13:24] <StevenK> DktrKranz: Still dealing with the eleventy million syncs
=== fargiola` is now known as fargiolas
[15:28] <superm1> tjaalton, well the ones i saw fly by were actually issues from jockey that are fixed by a jockey in -proposed, so it shouldn't be a big deal
=== winkle_ is now known as winkle
[17:00] <tjaalton> superm1: oh, excellent
[17:51] <jardi> hi all
[17:52] <jardi> I want to submit a patch to launchpad, but I can't remember the right diff options to make a "good" patch
[17:53] <Riddell> jardi: diff -urN
[17:53] <Riddell> the -u is the important bit
[17:55] <jardi> thanls Riddell
[17:55] <jardi> -l+k
=== nhandler_ is now known as nhandler
[18:46] <NCommander> cjwatson_, ping
=== lady_ is now known as lady
[19:31] <stgraber> tjaalton: hey, I just upgraded to the ati/readon driver you uploaded today (?), did you see bug 311867 ?
[19:31] <ubottu> Launchpad bug 311867 in xserver-xorg-video-ati "[M76 Mobility HD 2600] X fails to start - undefined symbol: exaDriverAlloc" [Undecided,New] https://launchpad.net/bugs/311867
[20:04] <tjaalton> stgraber: looks like another dupe of 311748
[20:07] <tjaalton> stgraber: or maybe not, but might be related
[20:09] <stgraber> tjaalton: hmm, in 311748 X doesn't seem to crash
[20:10] <tjaalton> stgraber: so in your case it returns to the console?
[20:10] <stgraber> tjaalton: yes
[20:11] <stgraber> tjaalton: well, it gets stuck half-initialized so I have to start X from ssh but starting it shows that undefined symbol: exaDriverAlloc and exit
[20:12] <tjaalton> stgraber: ok, sound odd since I can't see any dupes :)
[20:13] <stgraber> tjaalton: well, the error message sounds like a linking issue (that's why I then mentioned that I'm on amd64) but it's clearly the switch to EXA that introduced the bug
[20:16] <tjaalton> stgraber: ok, try building the driver without patch 104 and see if it helps
[20:19] <stgraber> building it now
[20:20] <stgraber> tjaalton: X starts just fine without 104
[20:24] <tjaalton> stgraber: ok then.. what if you force EXA from the conffile?
[20:25] <stgraber> started just fine
[20:25] <stgraber> (I added "Option" "AccelMethod" "EXA" to the Device section)
[20:25] <tjaalton> so the patch is broken.. let's wait until bryce get's back :)
[20:26] <stgraber> root@castiana:~/tmp1# grep EXA /var/log/Xorg.0.log
[20:26] <stgraber> (**) RADEON(0): Option "AccelMethod" "EXA"
[20:26] <stgraber> that's weird, I'd have thought it'd be more verbose and mention EXA more than one time
[20:31] <Nafallo> Baby!!! \o/
[20:31] <Baby> hi Nafallo :)))
[20:51] <tjaalton> bryce: stgraber confirmed that without ati patch 104 his X can start (bug 311867
[20:51] <ubottu> Launchpad bug 311867 in xserver-xorg-video-ati "[M76 Mobility HD 2600] X fails to start - undefined symbol: exaDriverAlloc" [Undecided,New] https://launchpad.net/bugs/311867
[20:53] <tjaalton> bryce: but forcing EXA by using AccelMethod in xorg.conf worked fine
[20:54] <tjaalton> (huh, thought I was on #ubuntu-x)
[20:55] <tjaalton> afk, BSG ->
[21:34] <YokoZar> Popcon says Wine has more installs than sun java6 now
[21:35] <YokoZar> (~39%)
[21:43] <YokoZar> Is the new Gecko backwards compatible?
[21:43] <YokoZar> oops wrong channel
[21:48] <directhex> YokoZar, the number doesn't take into account the number of people using sin java for windows with wine!
[21:52] <YokoZar> directhex: fair enough ;)
[21:53] <directhex> hm. "sin java". typo or freudian slip?
[21:53] <YokoZar> directhex: also doesn't count people using wine/firefox for windows flash
[21:55] <directhex> cool kids use wine/msie for windows silverlight!
[22:11] <NCommander> o_o;
[22:11] <NCommander> directhex, IT BURNS
[22:11]  * NCommander has WINE installed for the Windows version of firefox ...
[22:12]  * directhex burns NCommander 
[22:12] <directhex> NCommander, progress with ikvm! it compiles! and runs!
[23:31] <fta> stgraber, http://paste.ubuntu.com/98620/
[23:38] <stgraber> fta: thanks, applied. I don't really know why I changed it to parseString in commit 48 but it can't be a good reason (otherwise it'd be in the commit message) :)
[23:40] <fta> stgraber, thanks. i read you wanted to release 0.11 about a month ago, i can't see it in jaunty. am i missing something?
[23:40] <stgraber> fta: well, it's released. I'm just waiting on Debian to package it, then request the sync.
[23:41] <stgraber> fta: if it's not soon, I'll just upload the new upstream myself.
[23:41] <ebroder> Any core devs wiling to sponsor an upload for LP #216761 ?
[23:41] <ubottu> Launchpad bug 216761 in xen-3.3 "errors in xendomains init script" [Undecided,Confirmed] https://launchpad.net/bugs/216761
[23:42] <stgraber> fta: or do a 0.11.1 with the few patches I received and then upload it :)
[23:43] <fta> stgraber, and chance you can add a default action instead of showing usage ? i'm sick of adding "-i -"
[23:44] <fta> stgraber, especially that it's not consistent... you can add "-i -" or "-f text" to work around the test. imho, it's useless.
[23:47] <fta> (s/and/any)