> You are not creating software with the exact rules that I have made up in my head on how software should be made, so I'm going to build a wall and hate you and the code you make.
The funny thing about this comment is out of the gate it screams the exact reason people don't want the LLM generated code in their zone.
This reads similar to people who make prompt art getting upset that they're not taken seriously. People demand respect for something they didn't earn. Programmers who have worked hard for their knowledge don't want to listen to the person who thinks they know everything because the magical machine spit out some working code. It doesn't matter if they're right or wrong. They're annoying and entitled.
I'll just paste the classic Ian Malcom Jurassic Park quote.
> I'll tell you the problem with the scientific power that you're using here, it didn't require any discipline to attain it. You read what others had done and you took the next step. You didn't earn the knowledge for yourselves, so you don't take any responsibility for it. You stood on the shoulders of geniuses to accomplish something as fast as you could, and before you even knew what you had, you patented it, and packaged it, and slapped it on a plastic lunchbox
The argument about entitlement is silly: every generation just "gets" all the benefits and hard work of the previous generations. We didn't have to invent the assembly line, or electricity, or radios, or transistors...I can work for a few hours in a McDonalds and make enough to buy a cell phone that would have power unimaginable to someone even 50 years ago.
And heck, I'm a guy in my late 40s who's been programming since I was 8. I have a masters in CS, and have been a programmer for decades and I don't have the tiniest bit of resentment for the AI builders out there. Sure, they don't have my expertise, but they can still make cool stuff if they work at it.
What's really happening is that Codeberg was build on the assumption that the volume of code submitted would be small, and that assumption no longer holds. So now everyone is backing into reasons why limitations have to be imposed, and of course they won't say "It's because we built this whole thing assuming not that many people would use it, and now we have tons of use, so we're going to ban projects that are the most voluminous."
So instead, it's about malicious code, or copyright status, or entitlement, or some other topic.
As for the Jeff Goldblum quote: "You read what others had done and you took the next step" is literally how human progress is made.
I can see the point you are trying to make, and I'll make a counterpoint to that. When you play the "didn't earn" card, you are really getting nowhere. It's a bad argument.
It's putting up the gate, "I feel you didn't work hard enough for X result, but I did, so I must be better"
I drive a manual car. Should I hate every automatic car driver?
I have a master's degree in cybersecurity; should I shit on everyone who doesn't when they try to mention anything involving cybersecurity? I mean, I have a piece of paper that says I've earned it?
On some level, yes. I would listen to the advice of a doctor over a random person on the internet.
On another level, it's like saying, " Oh, you didn't have a blacksmith handcraft that knife for you, and you bought some machine-made knife? We should ban machine-made knives; the person making a knife using a machine didn't 'earn it '."
Perhaps you shouldn't create code; after all, you probably didn't develop a compiler. Or the computer chip running it... Or the OS.. Did you really earn the right to develop something?
It’s a big tent and anyone is welcome in. I learned to code on my own thanks to the efforts of people to write manuals, articles, text files on BBSs and NNTP. That’s just one way.
But try the other shoe on and walk a mile. You like writing code and solving problems yourself. Everyone in the world is telling you you’re irrelevant and should start using the machine or get left behind. The future is here and you’re not a part of it unless you give up and join us. Constantly.
It was hard before LLMs to convince software developers, let alone managers and product people and everyone else, to take quality seriously. Now it’s orders of magnitude harder. Because anything you say is met with someone who doesn’t care and thinks it’s all irrelevant. The machine will get better. It will solve all of these problems. Give up.
Fortunately that hasn’t deterred craftspeople in other trades who made knives, custom cabinetry, boats, houses and the like. Sure automation and progress has made a lot of knives available. Maybe more than we need. And most of them are disposable crap. But there are still knife makers.
All this talk of being reasonable is hard to take seriously when neither side will consider the other.
It’s also possible that even after considering all of the technical features and capabilities that LLMs bring to software development that some folks don’t find the cost of using them to be worth it. So they can say, “Sure, I will use such tools when and if I deem it appropriate.” Which may be never.
But we’re in a world right now where you can be coerced into feeling like there’s no choice but to accept those trade offs and just use the LLM.
I think it’s interesting and good that there exist places for people who share similar goals. There are few places for people who want to just do programming to chill out and not be surrounded by LLM generated code.
Also the scraping situation is totally getting out of hand. Having to keep up services is not easy or fun work. It’s just toil created at the expense of many for the benefit of few.
It's less about "earning" and more the attitude and approach to the work. The flippant nature of people expecting the same respect for some piece of work that was churned out by a machine, as someone who understands every part. I know I used "earn" a lot in the last reply, but that was a shorthand for something more complex. A combination of appreciation, discipline, and a common work style.
The LLM stuff is being gatekept because the people inside the gate care, and the people bringing the LLM code don't.
I drive a manual car too, and I get upset when someone who doesn't care about where they're putting their car, let alone the transmission, endangers me on the road. They're apathetic.
I work with people who have fully adopted claude in their work flows. They don't care about the code they push. They generate, prune, generate, prune in 10kloc increments. They're apathetic to others who might need to use this code and expect some kind of consistent architecture/interface. Its exhausting to deal with yet they love to say "I did this". I respect someone who tells me "I slopped this out" more, as they respect me enough to set expectations.
I think if you're trying to build a community of people who care, you must defend against those who are apathetic, or else they'll bring the whole community down.
This isn't to say all LLM users don't care about their work, but that there are far more people who are happy to push slop and take credit than the reverse. To carefully screen those people brings us back to the original complaint, LLMs turn programmers into paperwork drones. Attempting to screen these types of users takes away even more time from the purpose of the community.
> The flippant nature of people expecting the same respect for some piece of work that was churned out by a machine, as someone who understands every part.
> with people who have fully adopted claude in their work flows. They don't care about the code they push.
This just isn't true and it's the cynnical self-limiting assumption the belies rules like the TFA. It's like when HN was full of people claiming that you weren't a real engineer and didn't care about code if you used Javascript.
I've vibe-coded some pretty cool things that I use daily, and I've put some serious work into vibe-coding them. Months of work, infinite polish, infinite revisions, without writing a single line of code. It feels like the right level to be at with 20 years of software experience.
LLM use has merely freed up the chore work so that I can focus on more important directional concerns, polish the testing system to completely derisk it, and then do it in parallel with everything else I'm working on.
I'm starting to see the "LLM users don't care" assumption as a somewhat understandable defensive posture that mostly comes from resistance to the world changing and watching the value of our expertise go to zero. Something I still grapple with every week btw. Maybe if you accepted the idea that veteran software engineers can completely replace the code writing experience with AI while delivering good software, it involves swallowing a pill you're not ready to swallow?
Codeberg's new rule about LLMs is about as enforceable as a rule against bad code quality and that is the more interesting discussion.
Thank you; that is probably a better way to put into words what I am feeling.
I've never called myself a developer. But I've been playing with Linux and looking at code for the better part of 20 years.
I've done a lot of larger solution development. "Glue code," etc.
I think I use LLMs quite well. I mean, on some level, I understand software architecture and system architecture.
I've (before LLMs) had a master's degree, CISSP, Sec+, CEH. I am a pretty "qualified" person for whatever that means.
AI has made the chore work easy. It also allows me to think deeper about something than just bashing my head against code. Instead of spending time writing code, I'm researching technologies and thinking strategically about what something should look like years from now.
I've spent weeks and months, even on pet projects with LLMs. Things I use daily too.
But somehow "LLM coders don't care" is just annoying.
And I'd just grant the fact that LLM enables more low quality, low effort spam. Just like how opening up the internet to more people instead of just the research/academic nerds led to the same thing.
But the problem is a low quality spam problem, not a "people use LLMs at all" nor a "people don't care about the code like real artisanal engineers" problem. Neither of which are something you can police btw since they can't be proven.
How do you build a platform that is resistant to spam and low quality submissions? If you want an honor system, why not set the self-policing bar at low quality content instead of what you feel is a proxy for it?
> I've put some serious work into vibe-coding them. Months of work, infinite polish, infinite revisions, without writing a single line of code.
Not being pedantic, but this is not "vibe-coding", as other commenter and blog post means it at least. Parent emphasize lack of work, polish, revisions:
> They don't care about the code they push. They generate, prune, generate, prune in 10kloc increments. They're apathetic to others who might need to use this code and expect some kind of consistent architecture/interface.
> This isn't to say all LLM users don't care about their work, but that there are far more people who are happy to push slop and take credit than the reverse.
No, the TFA policy is specifically against LLM usage, not low quality code.
What you quoted is just how people paint LLM-heavy users just like how people used to paint JavaScript users as incompetent.
That there are people using LLMs to write 100% of their robust, derisked code disarms their argument or commits them to a sillier one like TFA's "we ban LLMs because you might get it to build a CI pipeline before you have users".
Hey, thanks for taking the time to reply. And you bring up some good points. I'll be the first person to tell someone I used Claude on this. Hell I make sure it says co-authored by claude on my github projects specifically because I'm not trying to hide the fact that I'm using LLMs.
I think you are correct, is not not really an LLM problem as much as its a person problem.
But there doesn't seem to be a good way to fix this. If anything LLMs just expose someone's integrity? I don't know, its a hard answer. Some people use LLMs just because they are translating languages.
The shift here is that up until now output (code, in this case) required a degree of competence. As such, we socialized that being able to put out the code showed a degree of competence. The act of putting together a PR with working tests showed you had a level of knowledge past "I want X to do Y."
Now, you have a lot of people taught that output (a working PR) is competence. They walk around thinking they are competent because they can produce output.
When faced with criticism on working code, they collapse and flail immediately because to admit they can't explain or defend the code is to admit they are not competent. From there you've hit pretty much all of the problems that the rest of the Internet suffers from: cognitive dissonance, digging in because being wrong threatens your constructed identity, constructing echo chambers of AI Influencers peddling the latest prompt, and so on.
Someone using an LLM to code isn't calling themselves a great coder just because the LLM is writing good code. Being a software engineer/developer (whatever you want to call it) is not defined solely on your ability to write code.
To go back to your point, before the F1 driver (LLM) was hired the owner was the driver, mechanic, and everything else. Now that they have a driver they can focus on all the other roles and just ensure the driver is not crashing.
It's a tragedy of the commons problem - if you allow LLM code, you get an infinite sewer of shitty commits from people who don't know you or care about you - so you close the door to everything.
Nobody wants to admit that we're going to have to go back to personal, verified relationships (it's who you know, not what you do) to fix it, unless we come up with something new.
Same problem with Eternal September - there was an "artificial" gateway that is now gone, but the need for some form of gateway still exists.
Developers who use LLMs: You are telling me that you have moral/ethical/practical issues with how I work but I'm going to ignore what you say to justify how I'm not at fault.
The moral/ethical high ground argument just spirals.
Oh you drive a car? You should justify that to me.
Oh you bought any piece of clothing, tech, etc where it was made in countries with cheap labor and bad labor laws? You should justify that to me.
Oh you use electricity? Idk some of that is made via non-renewable resources.
Oh you eat meat? So much animal suffering and land is being wasted.
I get using an LLM has some problems. But I find it ironic that for most people, they didn't have a problem with any of this until it started to affect their job and their expertise.. And their paycheck.
Where was your outcry for anything else? Or is it just performative?
I sure hope you are also only using products and services derived from union labor.
This is about the kind of code allowed in a place to store code. Your comparisons are obviously nonsensical in relation to the actual topic at hand. If you're a part of the transit community, they have plenty to say about car dependence and its ethics. If you're a part of the fashion community, they have loads to say on fast fashion and its ethics. The list goes on and on. For all them it seems totally reasonable for them to have their own ethical lines within their relevant spaces and communities. But it should be pretty obvious that none them have anything at all to do with Codeberg, or the general discussion of ethics at hand here. I don't see how you have any ability to claim what causes are issues someone cares about or prioritize based on them raising an issue relevant to the actual subject matter.
You're missing the point. I was only saying those things to show that arguing ethics and morals is bad because it's a never-ending self-imploding purity spiral. I am not trying to argue those views, nor am I refuting some argument because the replier is trying to claim some moral high ground.
If someone hangs a 'no cars' sign at the entrance to their private nature path you're not going to think they're saying 'You are not traveling according to the exact rules I have made up in my head about how one should travel, so I'm going to build a wall and hate you'
No, they don't want cars on their path, because cars would be harmful to the environment they're trying to foster.
I'm not asking you to justify your choices. What I'm saying is that you sound like you're deliberately ignoring what other people think as a precondition to justify your choices to yourself.
The changes are more nuanced. It's not banning AI assisted coding. It's banning purely and mostly vibe-coded projects. If you're using LLM/AI assistance responsibly, you're on the clear.
There's nuance, and Codeberg guys understand it, AFAICS.
I think I've avoided all of the LLM debate thus far, but I figure that some of my actual positions probably ought to be represented in this discussion instead of shadow-boxing.
I hate LLMs because I've had to sink a lot of hours into defending services and content that I host for myself or for others from companies that couldn't be bothered to write polite crawlers. This entirely unnecessary problem has contributed to some friction between me and a few clients as I try desperately to feather the pedal on traffic. This has subsided somewhat recently but the pain of it is still fresh.
I hate LLMs because they have taken a lot of goodwill and free labor from the creative community and turned it into a product. This isn't a new problem, it's been a point of sometimes rancorous argument in open source for decades, but this is another round of that old argument on what we owe or don't owe the people who donate labor and creativity to the commons.
I hate LLMs because I don't think Sam Altman or most of his social circle are good people and I don't want to give them more wealth and power. I believe Sam Altman is very comfortable enriching himself at the expense of humanity, and I want to still have some hope for the future of humanity. This position requires me to do whatever I can to deny any of my resources to Altman and his ilk, and to help foster the efforts of other people who want to do the same. To that end, I immediately support any protectionist efforts in the anti-LLM crowd.
I hate LLMs because they have inverted the effort it takes to generate some content versus consuming and considering that content. I am being drowned in a sea of garbage created by people in professional contexts who hand a few sparse notes to an LLM and ask it to generate a "professional email" or a "blog post" or a "technical document", and the quality on the overwhelming majority of this is terrible. It is having a severe and direct adverse effect on my ability to do work.
I hate LLMs because, after quietly observing differences between people who are at the far end of pro-LLM versus those at the far end of anti-LLM, I believe there's a consistent pattern of cognitive decline in the pro-LLM folks. I acknowledge this could just be me observing my own bias, but some literature is beginning to appear supporting it.
I hate LLMs because I love authenticity and creativity and constructive human endeavors, and LLMs are eating away at all of these and producing a homogeneous mass of superficial probabilistic crud. I have serious concerns about the future of art, creativity, and culture. They have been ever-changing and large parts of modern culture would be unrecognizable to someone 200 years ago, so this is more of a wait-and-see level of concern than a Big Hate, but so far, the Magic 8-Ball says, "Outlook not so good".
I hate LLMs because I'm strongly opposed to the idea that a few commercial companies, of whose finances I am deeply suspicious, should be made a required dependency in software development. This is an insane proposal to me. Even accounting for all of the upsides to LLM-driven development -- and I've followed all of that news just as much as all the negative takes -- it's nuts to me that there aren't more people worried about what this could mean for the future of software and technology. But, then again, I started my career during the height of Microsoft's domination of tech, so maybe that was formative for me.
I hate LLMs because I care about the environment that makes our existence possible and it's insane to me that so much of the tech sector suddenly stopped giving a shit about that. How can we, as a community, talk about the existential risk of climate change or the benefits of adopting renewable energy or cheer the development of ever-more-efficient power consumption while we also create a huge forcing function to convert as much electrical energy into thermal energy as possible? It's as if, practically overnight, everyone went from, "we need to do everything we can to reduce our demands on the environment or we're cooked" to "oooh, shiny new toy, and it only ruins everything around it, cool!"
I hate LLMs because I have been a hobbyist archivist for over a decade and a home computing enthusiast since the mid-1980s and right now I can't afford to buy a badly-needed hard drive or RAM or build out a new machine to even maybe experiment with an LLM in my own workshop because voodoo financing by everyone in the LLM ecosystem has totally wrecked the hobbyist computing economy, and there's no relief in sight.
I hate LLMs because I strongly suspect that the future they want is one of subscription computing and I will retreat back to retro computing before I go willingly into that long moonless night.
I hate LLMs because I am under enormous pressure, from many different angles, to also use them. I'm being asked -- required, in some cases -- to ignore all of my concerns about them and use them anyway because everyone else does and "obviously" they are the future. Because, somehow, all of my positions against LLMs make me unreasonable and a less valuable member of society.
LLMs could have been done in so many ways that I would have gladly and enthusiastically jumped right into. They are a really neat technology! I would love to be experimenting with them and playing with them and finding all the right ways to balance their benefits against their risks. But, I can't do that, because the way they've been developed requires me to set aside too many of the above concerns to be able to adopt them at all.
If I'm ever able to build a reasonable enough home system for the purpose, I might be able to, one day, change all the above to "I hate commercial LLMs...", but I have no idea right now how far away that day is, or if it will ever come. And, even that scenario would likely be a Faustian bargain, as it's likely that I'd be using open weights that were created under most of the same conditions as my concerns above.
I dislike current-gen AI for much the same reasons you mentioned: the externalities its development & use are causing.
The tech itself is neutral in this respect. It's those externalities that need to be addressed. Don't want to depend on a subscription service? Then run open models under your own control (if you 'have to' use AI). Don't want that new datacenter disturbing a town's water supply? Make 'em pay extra such that existing users can keep being served @ the same price.
Do such things across the board, and responsible AI companies will be forced to be more community-friendly.
I agree with most of what you write, and I think it is good that you mention these details. I also do not use LLM/generative-AI and I do not intend to use LLM/generative-AI, and many of my reasons are like what you have.
I hate LLMs because their pushers are hurrying us into the next financial crisis, which will impoverish another generation.
I hate LLMs because they have captured pretty much all R&D investment when we have far more important and urgent existential problems to solve. The world is literally burning in front of us, but no time for that because an even bigger model might make 0.03% less mistakes.
> You are not creating software with the exact rules that I have made up in my head on how software should be made, so I'm going to build a wall and hate you and the code you make.
More like:
> You are creating software using a tool incurring massive risks and costs for the rest of us even though we don't use it ourselves, we don't want you here
.
> plenty of people are using open offline models to code
…that needed to be trained and required siphoning the web too.
Personally, the reason I hate LLMs is not they trained on public code. They trained on code which doesn't allow that kind of thing to begin with.
Training LLMs is not the problem. Consent, violating copyleft licenses and other ethical issues is the problem.
IOW, I don't hate the tech. I hate how it's built. On top of infringements, being ignorant of others' labor and being proud of stealing what others done.
I’m sure that the people making this kind of argument would be quite pissed indeed if I trained an LLM on their proprietary code and used it for my commercial projects.
US Supreme Court has yet to make any explicit ruling though. It The EU top court (CJEU) is currently hearing arguments. I guess we will know in 2027-28 globally.
That's somewhat beside the point and bayindirh addressed this, but I'll answer to this bit:
> I'm going to open source my software so the world can use it for whatever they want!
Well, no, it's rarely "for whatever they want". Most open source licenses require attribution, and copyleft licenses require sharing any derivative work under the same license under some conditions. The terms of the licenses open source developer matter and are not there just to look pretty.
This idea that free software is released under no conditions needs to die. The few restrictions open source developers want for their code need to be respected!
And yes, I'm kinda mad the few lines of code I released to the world probably helped those despicable companies to train their despicable artifacts. We couldn't have possibly anticipated that this was coming, but now we know, and indeed the question of releasing publicly at all should be reconsidered. At the same time, we can't always stop doing anything that has bad consequences because otherwise we stop doing anything. So yes, I might both continue publishing free software and be mad at the same time. Any objection?
