Would be a terrible shame if lots of people opted out.

If you are a EU citizen you might also want to write a complaint to privacy@huggingface.co because they are collecting your personally identifiable information in machine-readable form which they are distributing to third parties.

you are viewing a single comment's thread
view the rest of the comments
[–] 159 points 2 days ago (5 children)

We want to give developers agency over their source code by letting them decide whether or not it should be used to develop and evaluate machine learning models.

fuck them seriously; if you want to do that then don't steal the repositories in the first place.

  • source
  • hideshow 10 child comments
  • [–] 109 points 2 days ago (2 children)

    Meanwhile at work we just had a training course that specifically said doing “opt out” instead of “opt in” violates the principle of informed consent.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 38 points 2 days ago (2 children)

    Kinda hope it uses my code. It's so terrible there's no doubt it will make the resulting code from the model worse even if the impact is miniscule

  • source
  • parent
  • hideshow 4 child comments
  • [–] 6 points 2 days ago (1 child)

    Unless I'm mistaken, this wasn't written by the folks that scaped GitHub in the first place, someone just wrote a small tool to semi-automate the process of searching the scraped dats, and submitting a GitHub issue to have it removed.

  • source
  • parent
  • hideshow 2 child comments
  • [–] -1 points 1 day ago (3 children)

    don’t steal the repositories

    How is this remotely stealing?

  • source
  • parent
  • hideshow 6 child comments
  • [–] 16 points 1 day ago (1 child)

    It's not, but it may be violation of licenses. And also, if it has personal information on it, that's probably illegal under the GDPR.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 0 points 1 day ago

    but it may be violation of licenses

    They excluded code with non-permissive licenses apparently:

    Each file is labelled permissive (at least one permissive license detected, no conflicting non-permissive license), no_license (no licenses detected, or only non-license legal texts such as CLAs), or non_permissive. The permissive allowlist follows the Blue Oak Council list plus licenses categorized as Permissive or Public Domain by ScanCode. Files classified as non_permissive are excluded from both released datasets.

    also, if it has personal information on it, that’s probably illegal under the GDPR.

    It's all public so I would be extremely surprised if that were the case.

  • source
  • parent
  • [–] 3 points 1 day ago* (1 child)

    the method on github is to fork it not steal all the data in an external storage for external purposes without consent of the user.

    edit: you can also report their whole account to github here: https://support.github.com/contact/report-abuse?category=report-abuse&report=bigcode-project&report_id=110470554&report_type=user

  • source
  • parent
  • hideshow 2 child comments
  • [–] 3 points 1 day ago (2 children)

    I'm pretty sure they just cloned the repos. That's how GitHub is designed to work. Have you never cloned a repo from GitHub?

  • source
  • parent
  • hideshow 4 child comments
  • [–] 5 points 1 day ago (1 child)

    Yes, but I used it under the licence terms of the original developer.

  • source
  • parent
  • hideshow 2 child comments
  • [–] -3 points 1 day ago (2 children)

    So have they. They filter by license - see my other comment.

  • source
  • parent
  • hideshow 4 child comments