Comment by Fraterkes
9 hours ago
I've been looking a lot at Godot (another big open source project) PRs lately, and there's been kind of a surge of wholy ai-generated PRs (both code and description). This is agains project-policy, so people creating these PRs usually get mildly told off. What's surprising is that while many submitters take that fairly well, some people get really indignant, essentially calling the maintainers ungrateful.
It's kinda surprising to me that even the people who are all in on ai haven't internalized that there's no inherent value in producing a big lump of code. They've massively decreased the work they put in but still expect the same pre-ai reaction/gratitude when submitting a big PR.
The pre-ai reaction was also unwarranted: committing a massive amount of potentially unmaintainable handwritten code isn't a necessarily positive contribution and any decent engineer (or person tbh) would understand that & not expect gratitude, no matter how concerted their effort.
In that context, I wouldn't expect an idiot (of which there has always been far too many in this industry) to change their behaviour in a post-ai world. They were always out of line & continue to be.
Fwiw, a non-technical employee in my workplace has begun submitting ai-generated prs to internal repos I maintain & they're of excellent quality, with review feedback graciously received & expediently addressed, so this isn't a matter of the idiots not being technical, it's an attitude problem.
Sure, but I think we should judiciously avoid the false equivalence yielded by only looking at this on a developer-by-developer basis, rather than systemically. The truth is that in practice, AI is not a neutral force. Obviously AI can enhance the output of smart, experienced developers and improve the efficiency of code reviews, mitigating the effects of garbage PRs. However, it increases the percentage of PRs contributed by entirely inexperienced and/or not-smart devs from zero to, potentially, the majority. It entirely removes the barriers inherent to coding that kept Dunning-Krueger cases from submitting ill-conceived or poorly constructed changes— actually getting them to run in some way, even poorly. That makes them much more difficult to distinguish from well-constructed PRs than those from, say, someone cargo-culting code from tutorials.
Moreover, as these tools become more expensive, people with money to blow on tokens will be able to drown maintainers that don’t have enough token-cash to help them deal with it. People see this as mostly a matter of time and energy, but I reckon it will soon be a financial issue.
I see AI as a barrier remover. Unfortunately some barriers are good or minimally necessary.
I think we'll need to revert to artificial barriers such as bonds, e.g., if you want to do a PR to my repository you need to pay a 10 dollar bond. If the PR is good and I want future PRs, you keep your bond. If it's slop and spam, I get 10 dollars for my time.
8 replies →
But your company employs said individual, whereas arbitrary drive-by patches from randos on open source projects with no consequence of submitting a mountain of garbage.
The answer: require a written proposal for changes before a patch will even be considered unless it is sufficiently small.
Also fight AI with AI: have a bot auto reject patches unless they can link to a previously approved enhancement document. Folks who commit minimal effort will f*ck right off.
Then the cognitive burden is focused on the ideas, and code authors should have at least conveyed the intent. If they actually care to invest their skin in the game then they need to collaborate and not just drop garbage on the front door.
just to clarify:
1. my company also employs its fair share of folk that would fit into the so-called "idiot" category of my post - I just thought it was of note that I have encountered exceptions to this stereotype
2. I fully support what Ladybird is doing here & find it unfortunate that they have to. I didn't intend my post to criticise their move - my example is definitely the rare exception in a sea of unmaintainable garbage. I do think however that it was already a challenging prospect to manage garbage oss prs in the pre-llm era (see umpteen posts on maintainer burnout) & I wouldn't have faulted any open source project for doing what ladybird is doing even pre-llm.
yes. github has very minimal controls for PRs (either shut them off completely, or not at all) but through GH webhooks you can essentially script "auto-close" of PRs that dont have correct preliminary approval.
it's time for there to be some really nice workflow tooling that people can plug into their GH repos that does this and other things.
> Fwiw, a non-technical employee in my workplace has begun submitting ai-generated prs to internal repos I maintain & they're of excellent quality, with review feedback graciously received & expediently addressed, so this isn't a matter of the idiots not being technical, it's an attitude problem.
It is hard for me to imagine another engineering discipline that would be totally fine accepting work from those who don't have the actual engineering background required to do the work.
If I had to push this take to the extreme: software engineers never learned class solidarity and it's now biting the industry in the ass.
Man, you must hate those handymen who put up YouTube videos showing how to do basic home maintenance. A truly class-conscious handyman would insist that the homeowner hire them to replace a light switch.
16 replies →
> hard for me to imagine another engineering discipline....
Well, that's already the case because you cant just call yourself an engineer and start signing off on projects. It's a legally protected title in a lot of places. You need a professional license, and can face legal liability for your decisions.
Software engineering is not engineering. Software craftmanship or even architecture would be a more accurate term. There are no devs that will go to prison if what they produce has, say, a major vulnerability. That alone disqualifies it from being engineering. There's no licensure, there's no liability, so already software development is not gatekept in any way like other engineering disciplines.
I mean, just go into an aerospace engineering office and say you want to move fast and break things, you'll get laughed out of the room.
No idea what you mean by class solidarity. There are only two; the capital owning class, and then everyone else (the working class). Most devs are working class just like everyone else.
Unless you're proposing that software should be gatekept to the level of other engineering disciplines?
2 replies →
Ooof. This is a big topic - I understand where you're coming from, but it's a common sentiment & one I've recently come to disagree more & more with.
Firstly: class solidarity. The apparent death of (or at least notable decline in) class solidarity is popularly lumped upon software engineers because they're relatively highly paid, but it's equally as absent in newly created positions (mainly within the IT sector) at all salary levels. There's been a concerted effort to erode class awareness in the private sector for the past 40+ years & it's been effective across all sectors, mostly in newly created job categories without pre-existing union culture. It's in no way specific to software engineering as a role nor to high salary positions.
Secondly: ai & llms. Currently these technologies are monopolised by corporate entities, with models generally being far too inefficient to democratise, so it's obviously tempting to conflate their very existence with their owners, but if you're singling out ai usage as some kind of affordance to the capitalist class you're missing the woods for the trees. You need to separate ownership from existence/usage.
3 replies →
happens all the time. Some business jerk outsources an entire initiative, forces it through review, and we get dumped with externally written crap we gotta deal with. So what if claude wrote it, actually claude is better than money wasted on those outsourced piles of crap projects
On this topic, "Open-Source Isn't About You": https://gist.github.com/richhickey/1563cddea1002958f96e7ba95...
Gratitude was maybe the wrong word. As the article mentions, before ai I think larger PRs, while sometimes inconsiderate, at least implied some amount of care / effort / good faith. In my experience, that was often rewarded with the maintainers at least taking a look at the code. I meant it's odd to have the same expectation when you dump 3000 lines in a pr that you won't even personally write a description for.
There is certainly a certain... entitlement? (It's not the perfect word, but I fail to find a proper term) from some of the vibe crowd. Like an attachment to the output and refusal to accept that most of the work was not theirs.
It is seen in the way they approach contributions but also in regular language. I created X, insistence that their 'curation' was very influencial to the output, difficulty to mention LLM contribution, attitude of 'I care about building while others lose time in details', refusal to engage with potential flaws, and so on.
It is surprisingly different to what I'm used to from senior devs, which behave like they always suspect their own work is flawed and half assed. Like impostor syndrome was reversed.
I've experienced the following sequence more than once at work, and I remain baffled by it each time:
- Receive a huge vibecoded PR for complicated new feature.
- Complain that this needs some design doc to figure out the right approach first.
- Author says no need for design doc, easier to have vibed implementation and discuss the concrete code instead of abstract document.
- I disagree (obviously), but review the PR with feedback along the lines: this entire approach is flawed, throw this out and start over.
- Author gets defensive, says "but this is already working and ready, let's just merge".
- I tell them there is no chance in hell this is getting merged. They go sulk to their manager that I'm not interested in helping them launch.
> helping them launch.
I think that's probably the key - sounds like you are at a place that rewards "launches" and not long term maintenance and so you are ruining their KPOs or promo packet or whatever.
1 reply →
I agree it's not "entitlement" specifically but there's something there. I guess by now everyone has experienced that type of person that "tries to help" by copy/pasting a bunch of AI slop and expecting you to work through the cognitive load of validating it.
The original post sums it pretty well, such big output inherently meant big effort, which was a proxy for good faith. Now that's gone.
It was a proof of work system. When work becomes cheap, it stops being proof.
> It is surprisingly different to what I'm used to from senior devs, which behave like they always suspect their own work is flawed and half assed
I never trust my own code. And one of the motivations of trying to be fluent with my editor, is to be able to quickly look at it when a bug is reported. I also don’t trust another person with their description of their code. Any surprise, and I’m looking at the source if it’s a available.
If a project has a rule to not submit AI generated PRs, people should never submit AI generated PRs to that project. It's spam. Or if the rule is more nuanced than that in relation to AI, it should be respected.
It's 100% an issue with the people with submitting these PRs. So, if someone has a history of having no issue with breaking project rules (let alone being arrogant about it), it should be a massive red flag about the person for any possible employer or future collaborator checking their profile, etc.
Why people are wilfully destroying their own reputation like that is beyond me. It's infinitely better to have no activity at all on your profile than to create a track record of bad behaviour.
I can empathize with this view but I've also seen one line clear bug fixes rejected because CC signed off on it.
There's no other correct fix - why do you care which pen I used to write it?
I can imagine in these cases the LLM is telling the "contributor" how smart they are and how much the project is loosing out, maybe saying something like: "It's not about maintaining project boundaries, it’s not about ensuring code quality; it’s a gatekeeping mechanism designed by traditionalists who feel threatened by forward-thinking creators like you who truly master the efficiency of AI."
Whether or not they are, I agree it is entirely possible to imagine them doing that, given what we know about how AI chat has reinforced people doing much worse to themselves and others.
(And the whole "miffed AI wrote a shitpost" thing)
I wouldn’t be surprised to find out this is part of their RHLF training, the attitude is so prevalent in these models.
really? in the way the parent comment describes? (genuine) i suppose i've never encountered a context where it would be relevant for an LLM to output that, interesting
> What's surprising is that while many submitters take that fairly well, some people get really indignant, essentially calling the maintainers ungrateful.
Why is it surprising that some people who expect results without spending any effort also feel entitled to receiving gratitude without putting in any thought?
If I don’t word these critiques in the most diplomatic way possible, this immediately turns into a discussion about the prevalence of anti-ai sentiment on hn. Which would be boring.
Since Godot got mentioned, I thought this recent talk by Juan Linietsky provides a reasonable alternative to heavily restricting contribution.
https://youtu.be/LDSwP37y_W4?si=fMIsdQ2yjoiGChb6
What would be great is something like Pangram for PRs and issues. Detects if the PR or issue is AI-generated (with multiple checkers for redundancy) so that maintainers can choose to auto-close and PR authors have to at least manually write _something_ before they launch 20k SLOC diffs.
> They've massively decreased the work they put in but still expect the same pre-ai reaction/gratitude when submitting a big PR.
I don't think that follows. They expect things to improve, they do something about it and might (unreasonably) be frustrated by what they think is a policy that stands in the way of quicker progress. The first part is certain, the second part less so, and the third is just speculation.
It's clear that open source project are struggling to understand what is going on and coming up with plans – like everyone else – but clearly there are better and worse ways to proceed in this new world, if popularity, adoption and progress are things you want to focus on.
One could call a big lump of code a Tumor.
> What's surprising is that while many submitters take that fairly well, some people get really indignant, essentially calling the maintainers ungrateful.
I would be more sympathetic if they actually spent meaningful time on these contributions and could maybe see an argument for wanting their work to be given due consideration (lots of caveats here), but from what I’ve seen that’s the exception rather than the rule with a lot of these case.
There massive value in AI PR's.
If a feature and ignored, it can forked to provide more value to the users.
If unaccepted bugfixes, the maintainers are just silly. They need to be forked off.
It's interesting to see this perspective in the wild. In the age of AI I wonder what "massive value" your PR is bringing to the maintainer. $1 worth of tokens?
As explained: New features. Bugfixes. Better analysis.
Only stoneagers would say that they are better than a good AI.
7 replies →
I mean, aren't you kind of proving the poster's point?
Fork away. If you want to put in the meaningful effort required to maintain and improve upon a project as significant as Godot, and feel that AI is a mechanism you want in order to do so, go for it. Clearly, the maintainers don't feel that that's the best approach to create the product they want to create, and they are not required to accede to the sense of entitlement of the community.
Even before AI it was trivial to setup a continuous merge script. I did that several years for several projects which refused my PR's.
Nowadays it's even more trivial.
And a community is more of a burden than an advantage nowadays. Users are ok, but a community not so. See python, perl, ruby, node and countless others.