crosswords / README.md
jeggers's picture
Update README.md
5c65f64 verified
metadata
license: odbl

Welcome to the Cryptic Crossword Clues dataset. This dataset was originally a project to practice webscraping and data processing skills but has since evolved to become a resource for cryptic crossword solvers and constructors globally.

About the dataset

The dataset comprises several cryptic crossword clues from published crosswords. It was created by George Ho as a personal side project. It has evolved to become a tool for cryptic crossword solvers and constructors, helping them understand how a clue or an answer has been used in the past.

Each row of the dataset represents one clue from a published cryptic crossword. The crossword grid itself is not saved. The dataset covers a large portion of scraped blogs and digital archives and is considered nearly exhaustive for the covered sources. Data points include the clue, answer, definition, puzzle_date, puzzle_name, source_url, and source. Note that some information may be missing from individual rows, introducing potential errors or noise in the data.

Fields included in the dataset

The Cryptic Crossword Clues dataset contains the following fields:

  • clue: This field contains the cryptic crossword clue from a published crossword. For example, 'Labourers going around spotted tools (8)'.

  • answer: This field consists of the answer to the stated cryptic crossword clue. For instance, for the clue 'Labourers going around spotted tools (8)', the answer would be 'HANDSAWS'.

  • definition: This field provides the definitions of the answers. In the above case, 'tools' would be the value of this field.

  • clue_number: This represents the number of the clue in the particular crossword it's sourced from, like '17a'.

  • puzzle_date: This field records the date on which the puzzle was published. It follows the format 'yyyy-mm-dd', as in '2017-08-25'.

  • puzzle_name: The name of the publication and/or puzzle from which the clue is taken is recorded in this field, such as 'Quick Cryptic 904'.

  • source_url: This holds the URL of the blog post from which the clue was scraped, such as 'https://times-xwd-times.livejournal.com/1799231.html'.

  • source: In this field, the blog from which the clue is sourced is stated. For instance, 'times_xwd_times'.

Please note that these are the only fields included in the published Cryptic Crossword Clues dataset due to copyright issues respecting the scraped blogs. However, the original dataset includes annotation, puzzle_url, is_reviewed, and datetime_reviewed fields. Users interested in accessing the full dataset for non-commercial use (such as academic research groups) are encouraged to reach out via email to the dataset creator.

License & Usage

This dataset is distributed under the Open Database License. Any rights in individual contents of the database are licensed under the Database Contents License. This allows for sharing and adapting the database, but requires users to attribute any public use of the database or works produces from the database.

Thus, while reuploading the dataset may be feasible, please ensure to attribute the original source and maintain the same open license to comply with the licenses.

Updates

The dataset is regularly updated to include newly published blog posts and crosswords. This process is done manually, with no set schedule. However, it's important to note that version control is not maintained, implying older versions of the dataset might be inaccessible.

If you intend to extend, augment, build on, or contribute to this dataset, please raise an issue on GitHub if you have a specific issue in mind.

Dataset Access

The dataset is published online using Datasette. For more information, please watch Simon Willison’s introduction to Datasette on YouTube.

Please respect the licensing and usage terms while using this dataset for your projects or research. If you have any queries or require any additional information, reach out to the curator of this dataset, George Ho.

For additional information regarding the dataset composition, collection process, cleaning, and distribution, be sure to check here.

Do let us know how you're using it! Enjoy working on those crossword clues!