Kernel Summit: "The time has come to make it harder for new people to contribute code to the kernel." -- Linus Torvalds reacting to AI assisted code contribution flood : linux@programming.dev

Kernel Summit: "The time has come to make it harder for new people to contribute code to the kernel." -- Linus Torvalds reacting to AI assisted code contribution flood (infosec.pub)

submitted 3 weeks ago by cm0002@infosec.pub to c/linux@programming.dev

48 comments fedilink hide all child comments

"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.

all 48 comments

sorted by: hot top controversial new old

[–] brucethemoose@lemmy.world 24 points 3 weeks ago* (last edited 3 weeks ago)

This is something OSS projects everywhere are dealing with, no matter what their previous policies were nor what their personal beliefs about LLMs are.

I can’t believe people here are acting like Linus and Linux “brought this on themselves.” Vibecoding spammers do not care about a project’s guidelines, nor their maintainers’ workload; they’re coming either way. And friction + stricter contribution standards are how other projects are dealing with this, too.

[–] illusionist@lemmy.zip 19 points 3 weeks ago (6 children)

suggestion of his was to create an AGENTS. md that makes generating kernel code expensive in compute and token time.

Why would one use the agents file if it makes it expensive?

hideshow 6 child comments

[–] Waterpumpee@lemmus.org 26 points 3 weeks ago (5 children)

I guess it states testing, simulating, double checking the submission. As should be done and would be done by humans. Instead of one shot guess in the wilds.

parent

hideshow 5 child comments

[–] illusionist@lemmy.zip 11 points 3 weeks ago

The quote sounds like a penalty. And your guess sounds like the minimum work that has to be done anyway. Completely different framing.

parent

[–] MagicShel@lemmy.zip -2 points 3 weeks ago (3 children)

"Penalty"? Fucking hell, that's exactly how my AI development is done. My personal tools repo that isn't shared with anyone has almost 2k tests that run every time there is a change. Also the documentation is updated, and multiple adversarial agents are run to challenge every decision from a security standpoint and a simplicity standpoint. And there's another guard I'm forgetting.

If you're going to code with AI, that's the least you can do.

parent

hideshow 3 child comments

[–] snooggums@piefed.world 4 points 3 weeks ago (2 children)

"My solution is to throw so many digital monkeys at the problem that they will write the works of Shakespeare."

parent

hideshow 2 child comments

[–] No_Thanks@nord.pub 2 points 3 weeks ago* (last edited 3 weeks ago)

Under capitalism, this token burn makes the shareholders climax at the expense of all else - datacenters are oligarch crack.

parent

[–] MagicShel@lemmy.zip 1 point 3 weeks ago

Shakespeare isn't the point — it works. It has saved countless hours of production support investigations, and has made it possible to answer questions and catch problems no one has been able to do in years because it would be prohibitively time consuming. I'm answering questions in a couple of minutes that would otherwise have just been a finger pointed at another team.

Why can't this guest log in? WAF problem because another team has a piss-poor page design? I would never have even thought of that. Forwards log entries to networking team for resolution.

parent

[–] justlemmyin@lemmy.world 36 points 3 weeks ago (2 children)

Is this the Linus who leaked the whole Linux Kernel source code on GitHub? Wow!

hideshow 2 child comments

[–] silver_wings_of_morning@feddit.dk 10 points 3 weeks ago (1 child)

Why is he not in jail yet?

parent

hideshow 1 child comment

[–] Swedneck@discuss.tchncs.de 5 points 3 weeks ago

he is, i mean obviously no one would willingly move to the US.

parent

[–] BigTrout75@lemmy.world 20 points 3 weeks ago

I get it, quality vs quantity

[–] onlinepersona@programming.dev 5 points 3 weeks ago

The kernel is already shit for humans to contribute to🤷

Let's be honest, mailing lists are garbage and just increase the barrier to contribution. They lack nearly everything useful for reviewing code and giving feedback.

They are wonderful for LLMs though. Somebody unwilling to learn the arcane crap needed to contribute just pushes it out to an LLM. It's easier for them to contribute than humans.

[–] OctopusNemeses@lemmy.world 12 points 3 weeks ago*

I've been sitting on a few patches from several years ago. I never submitted them because I don't want to go jump through the hoops again.

I get it though. Even before LLMs, the internet was already going down a path of slopification, with github repos being spammed with CV padder pull requests.

[–] nbsp@programming.dev 8 points 3 weeks ago* (last edited 3 weeks ago)

have not rtfa

maybe a simple referral system?

its going to make devs with connections a mega hot commodity $$$

edit: a future where sys admins get $$$ for know how to patch a kernel? more likely a future where we are stuck with shitty vendor vibe coded forked kernels with -1 chance of up streaming 🫠

[–] ICastFist@programming.dev 4 points 3 weeks ago

Well well well, if it isn't the effect everyone with a working brain predicted would happen by loosening up to AI.

It's also somewhat ironic that the "solution" to the bazaar is to become more like a cathedral - https://en.wikipedia.org/wiki/The_Cathedral_and_the_Bazaar

[–] anewfox@lemmy.zip 7 points 3 weeks ago* (last edited 3 weeks ago)

after decades on the internet, i ultimately agree with enforcing a skill floor when necessary. things were better when you had to kinda know a little bit about what you were doing to manage to get online.

[–] sanpo@sopuli.xyz 34 points 3 weeks ago (28 children)

Linus, buddy, you're the one that opened the floodgates by welcoming LLM code with open hands...

hideshow 28 child comments

[–] FishFace@piefed.social 42 points 3 weeks ago

But he didn't "welcome LLM code with open hands" really - he took a relatively neutral stance and said it needed to pass the same standards that human code does.

parent

[–] hexagonwin@lemmy.today 23 points 3 weeks ago

he didn't. it's more like him attempting to evaluate if it's any useful, and news websites suddenly sending out misleading headlines that linus is now 'vibecoding' or smth

parent

[–] kibiz0r@midwest.social 10 points 3 weeks ago* (last edited 3 weeks ago) (22 children)

Edit:

Okay my bad, I shouldn’t put words in his mouth. I’ll let him speak for himself.

Edit 2: Fixed grammatical ambiguity in the original

parent

hideshow 22 child comments

[–] FishFace@piefed.social 16 points 3 weeks ago (13 children)

But he isn't saying "damn people using AI", is he?

If you purposefully remove all nuance from what people say, you attribute to them a ridiculous position, but that's your problem not theirs.

parent

hideshow 13 child comments

[–] stembolts@programming.dev 9 points 3 weeks ago (1 child)

Lots of folks spend their time intentionally misunderstanding things so they can make jokes about the misunderstanding.

I don't get it, but you see it everywhere.

I think it's comforting for humans to believe things are a certain way, versus seeing them how they are.

parent

hideshow 1 child comment

[–] FishFace@piefed.social 4 points 3 weeks ago

As well as that I think people (nowadays? online? maybe it's just a general thing though) are just bad at accepting nuance or really imagining the positions of people which differ from their own.

parent

[–] kibiz0r@midwest.social 4 points 3 weeks ago (10 children)

I’m not a subscriber, so I could only see the summary — which mainly says it’s hard to summarize.

parent

hideshow 10 child comments

[–] FishFace@piefed.social 1 point 3 weeks ago (9 children)

So instead you've made up your own, consciously inaccurate summary. That sucks.

parent

hideshow 9 child comments

[–] kibiz0r@midwest.social 3 points 3 weeks ago (8 children)

Help me out then. What’s inaccurate? What would an accurate summary be?

We’ve got an earlier declaration from Linus on his stance towards AI, and a couple paragraphs about maintainers being overwhelmed by AI-assisted submissions, and an alleged statement from Linus that it’s time to make it harder to contribute.

I would argue that “let people use AI” is a fair summary of his earlier declaration, and that reports from overwhelmed maintainers do indicate that “people started using AI”.

So that leaves whether the current proposal of “making it harder to contribute” stems from “damn people using AI”.

Surrounding context in the summary certainly seems like there’s a causal link there, but if you have information that shows that the “harder to contribute” policy recommendation is not related to people using AI, I’d be very interested to learn more.

parent

hideshow 8 child comments

[–] FishFace@piefed.social 3 points 3 weeks ago (4 children)

Calling it "damning" people goes beyond the facts, and is antagonistic.

parent

hideshow 4 child comments

[–] kibiz0r@midwest.social 2 points 3 weeks ago* (last edited 3 weeks ago) (3 children)

That’s not the way I was using “damn” there, and that’s also not the usual usage of the meme format. I meant “damn” as an adjective. Like “those damn kids”. Not as a verb, like “damn those kids”.

It’s clear there’s some frustration, right? I think it’s fair to use “damn” in that way.

Edit:

This better?

parent

hideshow 3 child comments

[–] TrickDacy@lemmy.world 1 point 3 weeks ago* (last edited 3 weeks ago) (2 children)

Lol if you used a comma after the word damn, I think no one would've taken issue with your meme. It completely changes the meaning to leave it off! It also would've worked to put the comma after people. Which seems closer to what you meant.

parent

hideshow 2 child comments

[–] kibiz0r@midwest.social 2 points 3 weeks ago (1 child)

Well damn.

parent

hideshow 1 child comment

[–] TrickDacy@lemmy.world 0 points 3 weeks ago

Haha yeah, I never considered that it wasn't being used as a verb.

parent

[–] degen@midwest.social 2 points 3 weeks ago (2 children)

"AI can be used responsibly as long as people still use their brains"

People start using AI, stop using their brains

"Not like that. Stop that."

parent

hideshow 2 child comments

[–] kibiz0r@midwest.social 4 points 3 weeks ago (1 child)

That does seem like an accurate summary. And still fits perfectly in line with the stick-in-bike meme.

Or this:

parent

hideshow 1 child comment

[–] degen@midwest.social 2 points 3 weeks ago

That's the stance though, not the solution. Explicitly requiring more effort where it was previously implied would be a solution.

parent

[–] TrickDacy@lemmy.world 3 points 3 weeks ago* (last edited 3 weeks ago)

Just saw that edit. Lol. The idea behind it seems to still be ignoring nuance and implying that overreacting up front is the correct move rather than trying to avoid a rush to premature judgement.

Edit: In light of your other comment about how you meant the word damn, I think this was all a miscommunication.

parent

[–] Feyter@programming.dev 2 points 3 weeks ago

I'm completely against this opinion... But man that memes are funny. You earned your upvote. :D

parent

[–] teawrecks@sopuli.xyz 2 points 3 weeks ago

You understand the existing trusted parties contributing to the kernel are still allowed to use AI where they see fit, right?

He's been consistent and realistic about AI: LLMs are a powerful tool, smart people can make effective use of powerful tools in responsible ways, and irresponsible people can use the same tools to create a flood of noise. So create a system which allows smart, responsible people to make the kernel better, and prevents irresponsible people from making the kernel worse.

parent

[–] TrickDacy@lemmy.world 0 points 3 weeks ago* (4 children)

To call this a caricature and misinformation is an understatement. I have to think it's intentionally spreading lies.

I asked the person you responded to if they enjoy creating a nuanceless misinformation cartoon out of the truth and here you've gone and directly answered that with a resounding "yes".

parent

hideshow 4 child comments

[–] kibiz0r@midwest.social 3 points 3 weeks ago (3 children)

Help me out then. What’s inaccurate? What would an accurate summary be?

We’ve got an earlier declaration from Linus on his stance towards AI, and a couple paragraphs about maintainers being overwhelmed by AI-assisted submissions, and an alleged statement from Linus that it’s time to make it harder to contribute.

I would argue that “let people use AI” is a fair summary of his earlier declaration, and that reports from overwhelmed maintainers do indicate that “people started using AI”.

So that leaves whether the current proposal of “making it harder to contribute” stems from “damn people using AI”.

Surrounding context in the summary certainly seems like there’s a causal link there, but if you have information that shows that the “harder to contribute” policy recommendation is not related to people using AI, I’d be very interested to learn more.

parent

hideshow 3 child comments

[–] TrickDacy@lemmy.world 2 points 3 weeks ago* (last edited 3 weeks ago) (2 children)

"let people use ai" intentionally removes context. It implies he was good with ai slop in the kernel which he obviously never was and he never implied he was.

Then "damn them for using AI" isn't accurate in the least either. Taking steps to limit an issue isn't punitive or damning. Your image was a complete exaggeration in every way.

An accurate but non cartoonish version would be "hey let's not freak out about ai everyone, we don't need to overreact". Then later "we see a problem here so let's abate it". Your reaction to this seems to imply that if you ever make a statement that isn't a negative, then it's a positive. Then later if you ever present a stance that counters that perceived positive, then you're contradicting yourself. In reality even if the first stance was his, this "new" stance would represent learning and growing. You're discouraging that...

You pushed both statements in opposite directions in order to be mad, or something? I don't know, I don't get it, but it's hard to believe a person could honestly just even skim these articles and take away what you have.

parent

hideshow 2 child comments

[–] kibiz0r@midwest.social 3 points 3 weeks ago (1 child)

I just addressed this in the other thread, but I didn’t notice the ambiguity of “damn” in my original phrasing. I meant it in the sense of “those damn people using AI”, not “damn those people using AI”. It seemed obvious to me in my head, but tone doesn’t really carry across text that well.

Anyway, the point is not that he endorsed slop, but that he adopted a policy which people warned him would ensure slop arrived at his door on a regular basis.

Taking steps now that the problem has become unignorable is perfectly sensible. The silly part is that the problem was easy to predict from months away. He had a chance to get ahead of it. Instead, he chastised the people doing the warning.

parent

hideshow 1 child comment

[–] TrickDacy@lemmy.world 0 points 3 weeks ago

Yeah. You may have a good point that I know what you meant. I dunno all the details, I just got from the articles I mostly read that he'd discouraged freaking out about the issue and banning all LLM use. In any case I think the latest stance is a reasonable adjustment to reality. Maybe he should've had it earlier but I won't hold it against him too much

parent

[–] TrickDacy@lemmy.world 2 points 3 weeks ago (2 children)

That's not remotely what happened. Does it make you feel better to turn the truth into a cartoon with no nuance and obvious villains? I don't suspect you'll be argued into being reasonable, but people reading the comments should know that you are spreading lies.

Linus used an LLM to experiment in a branch. There is absolutely no evidence that he actually submitted any kernel code written by an LLM. He shouldn't have to hide his experiments with technology because some people are mindless drones who will become enraged and lie about what he's actually done.

parent

hideshow 2 child comments

[–] entwine@programming.dev 9 points 3 weeks ago (1 child)

There is absolutely no evidence that he actually submitted any kernel code written by an LLM.

The person you replied to never made such a claim.

Does it make you feel better to turn the truth into a cartoon with no nuance and obvious villains?

parent

hideshow 1 child comment

[–] TrickDacy@lemmy.world -2 points 3 weeks ago* (last edited 3 weeks ago)

Okay, true. But is there a better interpretation of "opened the floodgates by welcoming LLM code with open hands" than "look, he said he used it recently, so he demonstrated that he is fine with slop in the kernel"?

Trying to understand what the hell it would mean to say he "welcomed it with open hands" would mean and I assumed they incorrectly took him recently saying he experimented with llms to mean he pushed code written by it. Apparently like half the linux people here thought/assumed that.

parent

[–] cupcakezealot@piefed.blahaj.zone 3 points 3 weeks ago (1 child)

linux is what it is in spite of linus not because of him.

hideshow 1 child comment

[–] TrickDacy@lemmy.world 3 points 3 weeks ago

Yes, the man by far most responsible for a semblance of OS choice today is the real problem here because he tries to think objectively and avoid premature overreactions. Maybe you'd prefer an OS kernel that runs on outrage. Like every PR could be voted on in social media and the most upvoted gets merged and the least upvoted get posted on a wall of shame where the contributors are mercilessly mocked? Who needs decades of experience when you've skimmed some headlines that either confirm or deny your own worldview.

parent