“Torvalds said that the time has come to make it harder for new people to contribute code to the kernel. People like Kroah-Hartman have spent years trying to teach people how to contribute; it is time to stop doing that, he said.”

https://lwn.net/Articles/1099594/

It is difficult to summarize exactly how Linus got to this point. Most of the discussion at this years Summit appears to have been related to AI assisted code and how the maintainers are dealing with the increase in code submissions to LKML. Many of the maintainers expressed varying levels of frustration with AI assisted drive-by code submissions. Linus’ train of thought is that they make it intentionally more difficult for humans to contribute. A later suggestion of his was to create an AGENTS. md that makes generating kernel code expensive in compute and token time.

  • DonutsRMeh@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    7 days ago

    What’s he gonna do about code coming from corporations? We all know now they all vibe-code everything!

  • Lettuce eat lettuce@lemmy.ml
    link
    fedilink
    English
    arrow-up
    2
    ·
    8 days ago

    I was recently talking to a coworker about this same topic. I told them that one of my fears about LLMs is that the flood of vide-coded sloptributions would eventually make FOSS less free. Maintainers wouldn’t be able to keep up with the volume, and would resort to locking down contributions to well established community members and contributors.

    Basically, gone will be the days of skilled randos contributing freely to projects, or beginners jumping in to help with easy, small issues/fixes. Now you’ll need to be well known, be vetted formally by other contributors, etc.

    The agents.md compute sabotage idea is interesting, essentially Anubis but for your repository.

    • ryanpdg1@lemmy.ca
      link
      fedilink
      English
      arrow-up
      0
      ·
      7 days ago

      There might be a whole lot more forks out in the world 😖. Ultimately that’s what I’d do. Don’t want me to contribute? Fine! I’ll make my own with Blackjack and hookers!

      • thevoidzero@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        7 days ago

        You’re free to use a vibe coded linux kernel that you trust enough to run everything on your whole computer. I don’t

  • dustyData@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    8 days ago

    So, who gets the “I told you so” prize first?

    When he came out talking about allowing LLM code contributions everyone told him this would happen.

    • HaraldvonBlauzahn@feddit.org
      link
      fedilink
      English
      arrow-up
      0
      ·
      7 days ago

      When he came out talking about allowing LLM code contributions everyone told him this would happen.

      He has criticized the excessive hype around GenAI/LLMs long before. Torvalds has generally a very pragmatic approach, not a black-and-white one.

      Three things more:

      The Linux kernel is Torvalds project.

      The fact that experienced kernel developers can sometimes use LLMs to their advantage does not mean that the average Joe Codemonkey can do it. These algorithms are often described as a kind of competence multiplicator - but experience from education shows that they are more risky than that and can hinder learning.

      And third, we should in this question not put too much weight in the opinion of influential people. Behind the GenAI tech are huge corporations which pour billions of money into it. It is likely that they can pressure and bribe people and that they also pay influencers and bloggers to tout their stuff. Making decisions dependent on single influential persons, however competent and acclaimed they are, is therefore dangerous (see Eric Raymond, Eva Kaili or Gerhard Schröder). Torvalds is not Jesus, and I am pretty sure he didn’t apply for that role.

    • jimmy90@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      arrow-down
      2
      ·
      7 days ago

      they are still allowing LLM code that has been properly validated by a human

      “AI assisted drive-by code submissions” are still the problem

      nobody gets the luddite prize

    • Lettuce eat lettuce@lemmy.ml
      link
      fedilink
      English
      arrow-up
      1
      ·
      8 days ago

      He’s not perfect, obviously, he’s had some notable issues in the past. I think he’s a victim of his own optimism here.

      I think he genuinely didn’t believe so many people would just vibe code slop and spam maintainers with it.

      It’s kinda the classic, “I’ve always used XYZ responsibly and never had a problem. Doesn’t everybody else do that?” People do this all the time with credit cards, alcohol, gambling, extreme sports, etc. You think that because you’ve always done it “the right/smart/responsible way” it must be equally as obvious to everyone.

      • Lka1988@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        0
        arrow-down
        1
        ·
        7 days ago

        People tend to chill out as they get older. I’m way less tense now compared to 10 years ago, for a multitude of reasons. Trump’s bullshit isn’t helping, but overall my stress levels have gone down considerably.

        • lambalicious@lemmy.sdf.org
          link
          fedilink
          English
          arrow-up
          2
          ·
          4 days ago

          There’s a wide difference between being “chill” and letting yourself (and/or others) be raked without consideration.

          And, tbh, Linus is not even that old to be able to use that as an excuse. He was born in '69 - he’s barely older than I am. Eduardo Artés was born in 1951 and is actively fighting technofascism, fronting everything from social unrest to presidential campaigns.

    • Lka1988@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      7 days ago

      Hey at least he just came out and said it. No overly-sanitized corporate lingo statement, just “eh, I was wrong, so let’s figure out how to improve this process”.

      What’s the big deal?

      • ThotDragon@lemmy.blahaj.zone
        link
        fedilink
        English
        arrow-up
        1
        ·
        7 days ago

        Linus has shut down criticism of LLM generated code in the kernel and now it’s causing problems, as many people pointed out it would. Linus allowed and encouraged this and now he’s upset about the bed he made.

        • Lka1988@lemmy.dbzer0.com
          link
          fedilink
          English
          arrow-up
          0
          arrow-down
          1
          ·
          7 days ago

          Linus is very outspoken about his opinions. That should come as no surprise.

          That said, I don’t necessarily disagree with him here. There are far too many people who, at any mention of AI involvement with a codebase, immediately scream “SLOP” and “VIBECODED” without even trying to consider how it was used. And yes, there is no shortage of idiots who vibecode absolute slop, but there are also plenty of very talented devs who use it as just another tool in their toolbox, so to speak, and those are the people who will ensure they understand the code and make sure it works like it’s supposed to before committing.

    • jimmy90@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      arrow-down
      2
      ·
      7 days ago

      is this the anti AI version of “i’m tired boss”

      good luck with your personal kernel

      looooooooooool