Not surprising, but glad they pick a more sensible approach than other open source projects.

  • mmyu@lemmy.zip
    link
    fedilink
    arrow-up
    38
    ·
    5 days ago

    To me, this is the right approach. Contributors should not be submitting code that they don’t understand and can’t adjust.

    • nerdspice@lemmy.zip
      link
      fedilink
      arrow-up
      18
      ·
      5 days ago

      I agree. When I read their post, I thought good for them. It seems fair and they did a good job explaining the reasoning.

  • Alex@lemmy.ml
    link
    fedilink
    arrow-up
    14
    ·
    5 days ago

    It’s a hard problem. QEMU is currently reviewing it’s AI policy and we’re mindful to how helpful LLMs can be too someone new to the codebase while also capable of getting confused as the scope extends more widely over the repository.

    We’re open to wider changes being made with pre-approval but I suspect they will be special cases like guided re-factoring where they can remove the tedium from the engineer.

    We’ve been experimenting with using models for patch review but again it can be a time sync if the models go off on a wild adventure. If they could reliably catch the common gotchas that would at least alleviate some of the pressure on maintainers.

    The most important part of Godot’s policy is the humans responsibility and ownership of the code.

  • onlinepersona@programming.dev
    link
    fedilink
    arrow-up
    2
    arrow-down
    1
    ·
    5 days ago

    I don’t understand what concrete steps are being taken. People can still submit stuff and let all communication go through an LLM. What is concretely being changed here in the flow of things?

    • 0xDREADBEEF@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      6
      ·
      4 days ago

      If they see it they can reject the PR instead of going back and forth with a robot now without any complaints because they can just point to “follow our guidelines”

      Without the ability to say “It goes against the guidelines” the annoying robot can tirelessly complain and plead to let its robot code into the codebase. It can keep saying “there are no rules against robot slop contributions” or “there are no rules against using a robot as a medium for conversation” draining more time and energy from a finite pool of people capable of spending time and energy reviewing code and communicating.