‹ BackHN Continuity

Thread

Dutch governments builds alternative for Microsoft based on NixOS

1021 points · 583 comments · fjfaase

  1. ricardobeat · · focus · HN ↗
    A note to the developers in the project, if anyone is reading: your repositories are in violation of the new Codeberg policy that forbids AI-generated code [1].

    [1] <a href="https:&#x2F;&#x2F;blog.codeberg.org&#x2F;protecting-our-floss-commons-from-llms.html" rel="nofollow">https:&#x2F;&#x2F;blog.codeberg.org&#x2F;protecting-our-floss-commons-from-...

    1. mitxela · · focus · HN ↗
      Codeberg does not forbid AI-generated code. It forbids AI-generated projects. You would know this if you read the policy.
      1. llimllib · · focus · HN ↗
        The specific language is:

        &gt; projects that mostly consist of code written by &quot;generative AI&quot;-tools

        And &quot;mostly&quot; here seems egregiously undefined to me

        <a href="https:&#x2F;&#x2F;codeberg.org&#x2F;Codeberg&#x2F;org&#x2F;commit&#x2F;96fac426a32d1ba91ff879366d59bf1af54080c2" rel="nofollow">https:&#x2F;&#x2F;codeberg.org&#x2F;Codeberg&#x2F;org&#x2F;commit&#x2F;96fac426a32d1ba91ff...

        1. tcfhgj · · focus · HN ↗
          implies for me &gt;50% - instead of &gt;0% as suggested by @ricardobeat
        2. WarmWash · · focus · HN ↗
          The hardest part of policy isn&#x27;t coming up with smart policy, it&#x27;s figuring out where and how to draw the lines.
        3. mitxela · · focus · HN ↗
          What do you think &quot;mostly&quot; could mean?
          1. ricardobeat · · focus · HN ↗
            Take a peek at <a href="https:&#x2F;&#x2F;codeberg.org&#x2F;DAWO&#x2F;DAWO-Core&#x2F;commits&#x2F;branch&#x2F;main" rel="nofollow">https:&#x2F;&#x2F;codeberg.org&#x2F;DAWO&#x2F;DAWO-Core&#x2F;commits&#x2F;branch&#x2F;main

            Is this &quot;mostly developed with AI&quot; in your opinion? Not to mention the fact the policy leaves it open to interpretation is part of the problem...

        4. jenniferhooley · · focus · HN ↗
          As much as I hate vibe-coded &quot;slop&quot;. I feel like this is idiotic. I and most good engs with decades of experience do use AI now to type nearly all the actual &quot;code&quot;, but we read EVERY SINGLE line, always fix and rework every feature before a commit to make it clean, optimized, and de-slopified.

          But still, we wouldn&#x27;t be allowed to use codeberg because &quot;generative ai&quot; was used significantly for most lines of code?

          1. mitxela · · focus · HN ↗
            If it&#x27;s good, will they ever find out?
            1. jenniferhooley · · focus · HN ↗
              Maybe not... I personally wouldn&#x27;t use a service that would boot me off if they did find out because I spoke about it online or committed too much code one week or something.
          2. lowbloodsugar · · focus · HN ↗
            AFAICT they are dealing with resource exhaustion from slopcoding agents hammering them with copycat garbage. It&#x27;s existential. If you use AI to help you thoughtfully write code, review it with humans, and respect their service, I don&#x27;t think they will have a problem with you.
        5. ares623 · · focus · HN ↗
          &gt; And &quot;mostly&quot; here seems egregiously undefined to me

          and yet we seem to be convinced that English is the appropriate spec language of the modern times

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.