I can understand "untested AI-genned code is bad, and thus anything that reeks of AI is going to be scrutinized" - especially given that PostmarketOS deals a lot with kernel drivers for hardware. Notoriously low error margins. But they just had to go out of their way and make it ideological rather than pragmatic.
Having an LLM helps, especially when you're facing a new subsystem you're not familiar with, and trying to understand how things are done there. They still can't do the heavy duty driver work by themselves - but are good enough for basic guidance and boilerplate.
Do you genuinely think that people don’t know what documentation is? That’s insulting.
An LLM can help surface relevant information, taking your intent / goals into account, summarising vast quantities of code and indeed other documentation. That’s, like, their single most effective use.
My reading of their AI statement says your kernel contributions are no longer welcome in PostmarketOS, and also, since you're encouraging others in their space to use such tools, you're in violation of their code of conduct.
This applies to the person you're replying to too.
I think their policy is poorly thought out, and that little good will come of it. At best, it'll cause drama in the project, and discourage useful contributions. It's a shame, since we desperately need an alternative to the phone duopoly.
They explain why in their AI policy. It's an ethical stance. Of course they wouldn't notice if there aren't clear signs of LLM-ness, but that's not the main reason why they forbid it.
Thanks for the clarification. Not that I agree with their stance (the exact same could have been said at the start of the industrial revolution) but I respect it nonetheless.
> the exact same could have been said at the start of the industrial revolution
The pollution caused by said revolution is currently putting humanity at a serious risk of world war and maybe even extinction so... maybe they had a point? I'm not taking a strong stance either way here, but worth thinking about the downsides from the industrial revolution, too.
The AI policy linked from the OP explains why. It's half not wanting to deal with slop, and half ethical concerns which still apply when it's used judiciously.
If you’re unable to use your tools then that’s a skill issue on your part. Why are you so confident that something can’t be done just because you don’t do it? Some people are better than you at this particular skill. Learn to live with it.
The skill of cleaning up LLM-written slop to bring it to the human-like quality that any sane FLOSS maintainer would demand to begin with? It's just not worth it.
But at the same time I cannot imagine reverting to code with no help of LLMs. Asking stackoverflow and waiting for hours to get my question closed down instead of asking LLM? No way.
> However I'm not going back to asking stackoverflow and pretend that I have nowhere else to find answers.
That's not your only option.
What you're meant to do is understand the tools you're using well enough to not need to ask for help from anyone or anything else. Stack Overflow is useful, but it's a learning tool. If all you were doing before AI was copying and modifying other people's code, it's no wonder that you have taken to AI, because it's just a slightly more convenient form of that.
Once you get good enough at a programming language, you can code a lot from memory and logic. As in, you can think of a design and how to build it without having to look up someone else's code. It's still useful to keep notes to refer back to, and look up information online to jog your memory, but it's not always a question of finding other people's code to modify.
StackOverflow was also full of knowledgeable but objectionable people. I'm very glad not to have that energy in my life any more. Those that hate LLMs are welcome to continue using StackOverflow but I shan't be.
The main reason open source projects exist at all is because of people who started them with quite often fringe ideological leanings. Just look at the GNU project.
And fringe economical leanings, too. Just look at the GNU project: the firmware in printers is still of subpar quality, and GNU didn't really help to change that... and why on Earth would it, anyway?
It’s nowhere near 2003 anymore, and whether you or I like it or not there is a far greater visitation in ideology than there used to be. Your point is basically irrelevant.
There's still a line between values I disagree with and values that directly attack me as a person. The former is how many of us feel about some of our dependencies and most proprietary software we use, so it's clearly fine to some degree.
Yes and? Let's suppose your statement is 100% true, I genuinely don't see the point of these kinds of comments.
Why every time some person/group of people enact an anti-LLM policy in their project, other people feel the personal need to stress how useful LLMs are and how that project is bound to fail if they don't use it?
Postmarketos clearly exists and works, EVEN if LLMs were absolutely perfect for speeding up development ten folds, is there any absolute moral necessity to use them?
Also isn't this just moving the goalpost that LLM fanatics love to point out?
I'm pointing out that their expectation of AI-free OS is pointless.
Because AI-assisted code is most probably already present in devices they use.
And I dare say that even for PostmarktOS:
1) There's no way they can prevent AI-assisted code to reach their codebase.
2) They will most probably change this policy in the future lest other forks/projects outpace them in terms of utility and they get reduced to a carriage in a car world.
The stance is not to 'prevent AI-assisted code to reach their codebase.' It's not like AI-assisted code is literally poisonous and their codebase dies if touched.
The stance is to deter random vibe-coders trying to resume-max by submitting PRs to known open source projects. There are so many of them rn. Hopefully by making it clear (some of) them will realize doing that is just wasting their tokens.
> Postmarketos clearly exists and works, EVEN if LLMs were absolutely perfect for speeding up development ten folds, is there any absolute moral necessity to use them?
There's no moral necessity, but if you want to survive as a project moving forward, you'll have less and less velocity compared to projects using LLMs, so you'll eventually shrink and die as a project, because less people will contribute to a project that gets less features and bug fixes.
I don't understand why these projects have such a strong "moral" stance of "no AI ever", and instead they don't deploy LLMs to automatically review PRs based on their own guidelines, so that if the contribution is valuable, it gets through no matter if it was written by an LLM or not.
I don't work on open source stuff but I work at a financial institution and genai has been a huge productivity boost. I can easily write 2x - 5x more code than before genai.
Do you bring home 2x-5x more money every month then? Does your company make 2x 5x more profits?
The vibecoder paradox, everyone is 10x as productive, no one can show even a 1.2x increase in anything (besides bot generated comments, traffick and other background noise)
2x productivity will never result in 2x profit unless you somehow monopolize the productivity gain. Better tools can even result in negative profit gains. It's pretty much econ 101.
You say "uncompromising stance" with "justification", I say stubborn prejudice. They simply state the same weak, nonsensical complaints that apply to many other technologies that they undoubtedly don't have issues with and are happy with the use of.
I remember when people were crying about how much power a google search uses. This is the same thing all over again and it is as pointless now as it was back then.
> Google says it dropped the energy cost of AI queries by 33x in one year. The company claims that a text query now burns the equivalent of 9 seconds of TV.
No, it's entirely justified when quality of code matters. They don't want a thousand gallons of unreviewable slop. They want a reasonable amount of code that can be sensibility reviewed.
There are ways to achieve that without a blanket ban, if you read their AI policy it seems more "ethically" motivated. They certainly address this first, with many more words and 7 references.
They do go on to address code quality but it is more of an after thought with 0 references, less words and appears lower down the page.
Right now, their wiki page on device support [0] lists zero actual devices as "fully supported":
> These are the most supported devices, maintained by at least 2 people and have the functions you expect from the device running its normal OS, such as calling on a phone, working audio, and a functional UI.
> Besides QEMU devices, this is currently empty. The ports we had here earlier weren't as reliable as we would have liked. We plan to add new devices here with a higher standard.
The most recent smartphone in the Community section of that page is the Fairphone 4, released half a decade ago, in 2021. Pixel devices can trivially be bootloader unlocked, but that doesn't make the work that goes into supporting them much easier: the latest device in Testing is the 6a/6 Pro, from 2022, and its device page lists all the features but the most basic (touchscreen, flash, internal storage) as "Untested".
PostmarketOS doesn't exist in a vacuum. It’s the final stage of a device's life cycle. If the initial sales of new devices decline, the pool of available hardware for enthusiasts to tinker with in five years will be significantly smaller.
Yes. In five years, once the PMOS devs manage to get a 2025 device in working state, they might have less devices to play around with, so there could be an indirect effect on the project.
What I struggle to believe - what I don't believe - is that there any sort of connection between the report about likely declining sales and PMOS' announcement.
Please tell me more about how you used GPT-3 a few years ago and haven’t stopped blabbing about how bad it is ever since.
If you’re unable to look at a PR for a few minutes and glean that it’s not worth looking at, then that’s entirely a skill issue on your part. Don’t blame everyone else for what’s very clearly your own shortcoming.
If you’re finding a PR to be unreviewable, then reject it because it’s unreviewable, not by backwards-engineering some BS rule which results in you trying to control how people write their code.
I am completely confident that I could put together some LLM-assisted code that you could t distinguish from something else hand-written, with still enough LLM assistance to have been meaningfully beneficial.
There are many valid critiques of LLMs, but the whole “it’s banned because of code quality” approach is BS. This decision was clearly one rooted in the silly AI culture war.
This is all completely ignoring that the rule is completely unenforceable to begin with. It’s like the chuds on twitter talking about how they can “always tell” that someone is trans. It’s logically flawed.
Agreed. I would have chosen differently, but I appreciate the policy is unambiguous and explained succinctly with references.
Some people enjoy the outcome, others enjoy the process.
I find the criticism interesting. It's like one restaurant saying they'll use only electric stoves for the climate, then chefs all over the world calling them stupid naive for it.
It's like ethical arguments rationalizing local behavior are automatically interpreted as a global attack that has to be rejected.
Core Android functionality relies on eBPF in a way that PostmarketOS does not. PostmarketOS is much more of a linux distro than Android is. They are not very comparable.
AOSP patched kernels still include some features that are not in the mainline version. The LineageOS folks are working on support for mainline kernels, but AIUI it's not there yet.
> Submitting contributions fully or in part created by generative AI tools to postmarketOS.
So, autocomplete done by deterministic algorithms in IDEs are okay but autocomplete done by LLM algorithms - no, that's banned? Ok, surely everybody agrees with that, it's policy after all.
How it is possible to distinguish between the two in the vast majority of cases where the hand written code and autocompleted code is byte-by-byte identical?
Are we supposed to record video of us coding to show that we did type letters one by one?
> 2. Recommending generative AI tools to other community members for solving problems in the postmarketOS space.
Is searching for pieces of code considered parts of solving problems?
Then how do we distinguish between finding a a required function by grepping code or by asking LLM to search for it?
Can we ask LLM questions about postmarketOS? Like, "what is the proper way to query kernel for X given Z"?
If a community members asks this question and I already know the answer via LLM, then am I now banned from giving the correct answer?
--
Don't get me wrong. I am sick and tired of the vomit-inducing AI bullshit (as opposed to the tremendous help that LLMs provide to experienced devs).
I fail to see how a policy like this is even enforceable let alone productive and sane.
On the other hand, I absolutely see where is this policy coming from. It seems that projects are having a hard time navigating the issue and looking for ways to eliminate the insurmountable amount of incoming slop.
I think we still haven't found a right way to do it.
> So, autocomplete done by deterministic algorithms in IDEs are okay but autocomplete done by LLM algorithms - no, that's banned? Ok, surely everybody agrees with that, it's policy after all.
Because autocomplete still requires heavy user input and a SWE at the top of the decision making tree. You could argue that using Claude or Codex enables you to do the same thing, but there's no guarantee someone isn't vibecoding and then not testing adequately to ensure, firstly, that everything can be debugged, and secondly, that it fits in with the broader codebase before they try to merge or PR.
Plenty of people use Claude like an autocomplete or to bounce ideas off of, which I think is a great use case. But besides that, using a tool like that in more extreme ways is becoming increasingly normalized and probably not something you want in your codebase if you care about code quality and avoiding pointless bugs.
Every time I see a post on HN about some miracle work Claude did it's always been very underwhelming. Wow, it coded a kernel driver for out of date hardware! That doesn't do anything except turn a display on... great. Claude could probably help you write a driver in less time, but it'll only really work well, again, if you're at the top of the hierarchy of decision making and are manually reviewing code. No guarantees of that in the FOSS world because we don't have keyloggers installed on everybody's machine.
Yes, actually. Knowingly violating the policies of a project while pretending you aren't, so you can continue participating in the fully voluntary project, does make you a jerk.
If you don't like the policies they set, just leave.
I'm willing to bet that every single person on here complaining has zero contributions to PostmarketOS.
that ship has sailed with codex 5.3 in 90% SWE jobs, unfortunately. I expect the next 9% won't survive the following 12 months and the last 1% is done within 5 years.
it isn't even about principles - projects not using gen AI will become basically irrelevant, the pace of gen AI allowed competitors will be too great.
Alright, let's see Codex 5.3 create a competitor to postmarketOS (without just copying the homework of other devs). If you believe in the technology so much, put it to the test, see what it can really do.
Sure, AI has developed quickly, but let's see it take on a real engineering challenge, rather than regurgitating boilerplate code.
Writing device drivers from incomplete specs is much harder than "writing a whole application" where the specs are clearly defined and there's a lot more example code to reference. If you believe in AI so much, and believe that it's unreasonable for postmarketOS to not want to use it, put it to the test, prove the doubters wrong, what have you got to lose?
What does a developer who writes a driver from incomplete specs do? Writes some values in some registers, sees how the device behaves, updates the spec. Rinse and repeat. Sounds exactly the kind of stuff coding agents thrive at - a verifiable loop. And they can do it 24x7 until done.
Haha, are you trying to suggest you'll have lost much by putting an AI tool to the test? You seem to think it's powerful enough to do the work of porting Alpine Linux (or equivalent) to new hardware without human intervention (beyond the initial prompt), what exactly are you losing by trying this out? It's not your time, as you would have spent less time on giving a simple instruction to an AI tool than you spent in talking to me.
Perhaps the reality is that you know AI needs more hand-holding than this, and the tools aren't up to the task you're thinking of setting them.
You are also strangely fixated on today's capabilities, completely missing the exponential we are on.
In a few months will have posts here from device driver writers explaining how they hooked up a phone to an Arduino and a video camera and how the AI is automatically writing device drivers.
> You are also strangely fixated on today's capabilities
I am talking about today's capabilities because this comment thread started with the suggestion that the benefits of AI for coding was no longer avoidable after the launch of Codex 5.3.
> In a few months will have posts here from device driver writers explaining how they hooked up a phone to an Arduino and a video camera and how the AI is automatically writing device drivers.
A few months? Almost zero chance. If it happens in the next 5 years I'd be less surprised, but I suspect it'll take longer.
> Fun that you had to caveat it with some hand wavy homework bull.
Not really. If AI is just copying someone else's code, it's not really designing it is it. If you want it to truly design something, it needs to be designing it using the same constraints that the human engineers would face, which means it doesn't get the luxury of copying from others, it has to design things like device drivers with the same level of information that human engineers get (e.g. device specifications and information gathered through trial and error).
Are you suggesting that a human being writes an OS in a vacuum without seeing any other OS or looking into how it is built. That feels a little facetious, no?
> Are you suggesting that a human being writes an OS in a vacuum without seeing any other OS or looking into how it is built. That feels a little facetious, no?
No, I'm suggesting in order for it to be a fair test, you need to impose the same restrictions that a human engineer would face.
For example, consider the work done by the Nouveau team in building a set of open source GPU drivers for NVIDIA GPUs. When they started out the specs were not so widely available. They could look at how GPU drivers were developed for other GPUs, but that is not going to be a substitute for exploratory work. Let's see how well AI does at that exploratory work. I think you'll find it's a lot harder than common uses for AI today.
This sounds impractical and like they will probably not keep the ban
AI use should be able to accelerate the development of ports on currently unsupported or undersupported devices which would directly support the project
I guess I wouldn't worry about the policy, they will probably naturally switch it if / when AI becomes more useful in practice
There are some really baffling takes here. And it doesn't really matter how good or bad coding agents are.
Coding agents greatly reduce the barrier to contributing something that at least looks okay at the surface, so reviewing contributions will quickly become even more of a bottleneck. Manual contributions used to filter away most low effort attempts, or at least they could easily be identified and rejected.
That dynamic is now different and the maintainers risk being swarmed with low effort contributions, that will take a lot of time to review and respond to. Some AI contributions might be reviewed and revised and overall be of acceptable quality, but how can the maintainers know which without reviewing everything, good and bad alike.
I think we will see multiple attempts like this to shift things back to the old dynamic, by rejecting things that can be identified as AI generated a glance, but I suspect over time it will be difficult to do so, so my prediction is that we will soon see more open source repos stop accepting outside contributions entirely.
Even if LLMs one day will be good enough to quickly produce code that is on par with humans (which I strongly doubt), why would the contributors have any incentive to have someone else do that (the easy part), rather than just doing it themselves?
mono442 | 9 hours ago
MonkeyClub | 9 hours ago
ACCount37 | 8 hours ago
I can understand "untested AI-genned code is bad, and thus anything that reeks of AI is going to be scrutinized" - especially given that PostmarketOS deals a lot with kernel drivers for hardware. Notoriously low error margins. But they just had to go out of their way and make it ideological rather than pragmatic.
yehoshuapw | 8 hours ago
ACCount37 | 8 hours ago
Having an LLM helps, especially when you're facing a new subsystem you're not familiar with, and trying to understand how things are done there. They still can't do the heavy duty driver work by themselves - but are good enough for basic guidance and boilerplate.
trollbridge | 8 hours ago
UqWBcuFx6NV4r | 7 hours ago
Do you genuinely think that people don’t know what documentation is? That’s insulting.
An LLM can help surface relevant information, taking your intent / goals into account, summarising vast quantities of code and indeed other documentation. That’s, like, their single most effective use.
hedora | 8 hours ago
This applies to the person you're replying to too.
I think their policy is poorly thought out, and that little good will come of it. At best, it'll cause drama in the project, and discourage useful contributions. It's a shame, since we desperately need an alternative to the phone duopoly.
crimsonnoodle58 | 8 hours ago
If you use AI to make repetitive tasks less repetitive, and clean up any LLM-ness afterwards, would they notice or care?
I find blanket bans inhibitive, and reeks of fear of change, rather than a real substantive stance.
jonathrg | 8 hours ago
https://docs.postmarketos.org/policies-and-processes/develop...
crimsonnoodle58 | 8 hours ago
coldpie | 7 hours ago
The pollution caused by said revolution is currently putting humanity at a serious risk of world war and maybe even extinction so... maybe they had a point? I'm not taking a strong stance either way here, but worth thinking about the downsides from the industrial revolution, too.
jsheard | 8 hours ago
The AI policy linked from the OP explains why. It's half not wanting to deal with slop, and half ethical concerns which still apply when it's used judiciously.
zozbot234 | 8 hours ago
That never happens. It's actually easier to write the code from scratch and avoid LLMness altogether.
UqWBcuFx6NV4r | 7 hours ago
zozbot234 | 7 hours ago
egorfine | 7 hours ago
But at the same time I cannot imagine reverting to code with no help of LLMs. Asking stackoverflow and waiting for hours to get my question closed down instead of asking LLM? No way.
cuu508 | 7 hours ago
And doesn't that bother you a little?
If you listen to podcasts, check out this podcast episode: https://www.pushkin.fm/podcasts/cautionary-tales/flying-too-...
It is about Air France 447, but also draws parallels to AI and self-driving cars
egorfine | 6 hours ago
However I'm not going back to asking stackoverflow and pretend that I have nowhere else to find answers.
ZenoArrow | 4 hours ago
That's not your only option.
What you're meant to do is understand the tools you're using well enough to not need to ask for help from anyone or anything else. Stack Overflow is useful, but it's a learning tool. If all you were doing before AI was copying and modifying other people's code, it's no wonder that you have taken to AI, because it's just a slightly more convenient form of that.
egorfine | 4 hours ago
Aren't we all in a sense?
ZenoArrow | 34 minutes ago
Once you get good enough at a programming language, you can code a lot from memory and logic. As in, you can think of a design and how to build it without having to look up someone else's code. It's still useful to keep notes to refer back to, and look up information online to jog your memory, but it's not always a question of finding other people's code to modify.
fartfeatures | 3 hours ago
jonathrg | 8 hours ago
trollbridge | 8 hours ago
Joker_vD | 8 hours ago
UqWBcuFx6NV4r | 7 hours ago
Joker_vD | 8 hours ago
As long as they align with the correct (i.e. yours) values, of course. When they adopt the wrong values, it's not fine.
jonathrg | 8 hours ago
debugnik | 8 hours ago
xantronix | 7 hours ago
ForHackernews | 8 hours ago
hu3 | 8 hours ago
And I highly doubt iOS and Android are free from LLM assisted code at this point.
mpol | 8 hours ago
[OP] pantalaimon | 7 hours ago
hu3 | 7 hours ago
Not even humans can do that. Documentation needs to at least be reverse-engineered and understood before implementation.
fruitworks | 5 hours ago
fartfeatures | 7 hours ago
imadr | 8 hours ago
Why every time some person/group of people enact an anti-LLM policy in their project, other people feel the personal need to stress how useful LLMs are and how that project is bound to fail if they don't use it?
Postmarketos clearly exists and works, EVEN if LLMs were absolutely perfect for speeding up development ten folds, is there any absolute moral necessity to use them?
Also isn't this just moving the goalpost that LLM fanatics love to point out?
hu3 | 7 hours ago
Because AI-assisted code is most probably already present in devices they use.
And I dare say that even for PostmarktOS:
1) There's no way they can prevent AI-assisted code to reach their codebase.
2) They will most probably change this policy in the future lest other forks/projects outpace them in terms of utility and they get reduced to a carriage in a car world.
raincole | 7 hours ago
The stance is to deter random vibe-coders trying to resume-max by submitting PRs to known open source projects. There are so many of them rn. Hopefully by making it clear (some of) them will realize doing that is just wasting their tokens.
hu3 | 7 hours ago
But to be clear their AI instance is as clear-cut as can be. Their instance IS INDEED to "prevent AI-assisted code to reach their codebase".
> The following is not allowed in postmarketOS:
> Submitting contributions fully or in part created by generative AI tools to postmarketOS.
source: https://docs.postmarketos.org/policies-and-processes/develop...
plqbfbv | 6 hours ago
There's no moral necessity, but if you want to survive as a project moving forward, you'll have less and less velocity compared to projects using LLMs, so you'll eventually shrink and die as a project, because less people will contribute to a project that gets less features and bug fixes.
I don't understand why these projects have such a strong "moral" stance of "no AI ever", and instead they don't deploy LLMs to automatically review PRs based on their own guidelines, so that if the contribution is valuable, it gets through no matter if it was written by an LLM or not.
fruitworks | 5 hours ago
surgical_fire | 8 hours ago
Why don't you share the list of very useful things you created instead, mono442?
nananana9 | 8 hours ago
mono442 | 8 hours ago
jsheard | 8 hours ago
qsera | 8 hours ago
mono442 | 7 hours ago
qsera | 7 hours ago
hakube | 8 hours ago
lm28469 | 8 hours ago
The vibecoder paradox, everyone is 10x as productive, no one can show even a 1.2x increase in anything (besides bot generated comments, traffick and other background noise)
raincole | 5 hours ago
jonathrg | 9 hours ago
LaSombra | 8 hours ago
UqWBcuFx6NV4r | 7 hours ago
GaryBluto | 7 hours ago
fartfeatures | 7 hours ago
I remember when people were crying about how much power a google search uses. This is the same thing all over again and it is as pointless now as it was back then.
https://arstechnica.com/ai/2025/08/google-says-it-dropped-th...
> Google says it dropped the energy cost of AI queries by 33x in one year. The company claims that a text query now burns the equivalent of 9 seconds of TV.
idiotsecant | 7 hours ago
fartfeatures | 7 hours ago
They do go on to address code quality but it is more of an after thought with 0 references, less words and appears lower down the page.
The timing is also suspicious, shortly after publication of this report: https://www.reuters.com/business/media-telecom/smartphone-ma... which forecasts declining smartphone sales meaning less devices for this OS to run on.
mftrhu | 6 hours ago
Why would declining sales of new smartphones have anything to do with PostMarketOS, which only supports phones more than half a decade old?
zozbot234 | 6 hours ago
mftrhu | 6 hours ago
> These are the most supported devices, maintained by at least 2 people and have the functions you expect from the device running its normal OS, such as calling on a phone, working audio, and a functional UI.
> Besides QEMU devices, this is currently empty. The ports we had here earlier weren't as reliable as we would have liked. We plan to add new devices here with a higher standard.
The most recent smartphone in the Community section of that page is the Fairphone 4, released half a decade ago, in 2021. Pixel devices can trivially be bootloader unlocked, but that doesn't make the work that goes into supporting them much easier: the latest device in Testing is the 6a/6 Pro, from 2022, and its device page lists all the features but the most basic (touchscreen, flash, internal storage) as "Untested".
[0] https://wiki.postmarketos.org/wiki/Devices
fartfeatures | 3 hours ago
mftrhu | 2 hours ago
What I struggle to believe - what I don't believe - is that there any sort of connection between the report about likely declining sales and PMOS' announcement.
UqWBcuFx6NV4r | 7 hours ago
If you’re unable to look at a PR for a few minutes and glean that it’s not worth looking at, then that’s entirely a skill issue on your part. Don’t blame everyone else for what’s very clearly your own shortcoming.
If you’re finding a PR to be unreviewable, then reject it because it’s unreviewable, not by backwards-engineering some BS rule which results in you trying to control how people write their code.
I am completely confident that I could put together some LLM-assisted code that you could t distinguish from something else hand-written, with still enough LLM assistance to have been meaningfully beneficial.
There are many valid critiques of LLMs, but the whole “it’s banned because of code quality” approach is BS. This decision was clearly one rooted in the silly AI culture war.
This is all completely ignoring that the rule is completely unenforceable to begin with. It’s like the chuds on twitter talking about how they can “always tell” that someone is trans. It’s logically flawed.
kruffalon | 7 hours ago
That's like calling a person going for seconds a conservative (in the USA political sense).
randusername | 5 hours ago
Some people enjoy the outcome, others enjoy the process.
I find the criticism interesting. It's like one restaurant saying they'll use only electric stoves for the climate, then chefs all over the world calling them stupid naive for it.
It's like ethical arguments rationalizing local behavior are automatically interpreted as a global attack that has to be rejected.
chasil | 9 hours ago
9cb14c1ec0 | 8 hours ago
zozbot234 | 8 hours ago
egorfine | 7 hours ago
So, autocomplete done by deterministic algorithms in IDEs are okay but autocomplete done by LLM algorithms - no, that's banned? Ok, surely everybody agrees with that, it's policy after all.
How it is possible to distinguish between the two in the vast majority of cases where the hand written code and autocompleted code is byte-by-byte identical?
Are we supposed to record video of us coding to show that we did type letters one by one?
> 2. Recommending generative AI tools to other community members for solving problems in the postmarketOS space.
Is searching for pieces of code considered parts of solving problems?
Then how do we distinguish between finding a a required function by grepping code or by asking LLM to search for it?
Can we ask LLM questions about postmarketOS? Like, "what is the proper way to query kernel for X given Z"?
If a community members asks this question and I already know the answer via LLM, then am I now banned from giving the correct answer?
--
Don't get me wrong. I am sick and tired of the vomit-inducing AI bullshit (as opposed to the tremendous help that LLMs provide to experienced devs).
I fail to see how a policy like this is even enforceable let alone productive and sane.
On the other hand, I absolutely see where is this policy coming from. It seems that projects are having a hard time navigating the issue and looking for ways to eliminate the insurmountable amount of incoming slop.
I think we still haven't found a right way to do it.
kunai | 7 hours ago
Because autocomplete still requires heavy user input and a SWE at the top of the decision making tree. You could argue that using Claude or Codex enables you to do the same thing, but there's no guarantee someone isn't vibecoding and then not testing adequately to ensure, firstly, that everything can be debugged, and secondly, that it fits in with the broader codebase before they try to merge or PR.
Plenty of people use Claude like an autocomplete or to bounce ideas off of, which I think is a great use case. But besides that, using a tool like that in more extreme ways is becoming increasingly normalized and probably not something you want in your codebase if you care about code quality and avoiding pointless bugs.
Every time I see a post on HN about some miracle work Claude did it's always been very underwhelming. Wow, it coded a kernel driver for out of date hardware! That doesn't do anything except turn a display on... great. Claude could probably help you write a driver in less time, but it'll only really work well, again, if you're at the top of the hierarchy of decision making and are manually reviewing code. No guarantees of that in the FOSS world because we don't have keyloggers installed on everybody's machine.
egorfine | 7 hours ago
But again: how do we distinguish between manual code input and sophisticated autocomplete?
idiotsecant | 7 hours ago
egorfine | 7 hours ago
Blackthorn | 6 hours ago
If you don't like the policies they set, just leave.
I'm willing to bet that every single person on here complaining has zero contributions to PostmarketOS.
aboardRat4 | 7 hours ago
egorfine | 7 hours ago
baq | 7 hours ago
that ship has sailed with codex 5.3 in 90% SWE jobs, unfortunately. I expect the next 9% won't survive the following 12 months and the last 1% is done within 5 years.
it isn't even about principles - projects not using gen AI will become basically irrelevant, the pace of gen AI allowed competitors will be too great.
ZenoArrow | 7 hours ago
dist-epoch | 7 hours ago
ZenoArrow | 7 hours ago
Writing device drivers from incomplete specs is much harder than "writing a whole application" where the specs are clearly defined and there's a lot more example code to reference. If you believe in AI so much, and believe that it's unreasonable for postmarketOS to not want to use it, put it to the test, prove the doubters wrong, what have you got to lose?
dist-epoch | 7 hours ago
What does a developer who writes a driver from incomplete specs do? Writes some values in some registers, sees how the device behaves, updates the spec. Rinse and repeat. Sounds exactly the kind of stuff coding agents thrive at - a verifiable loop. And they can do it 24x7 until done.
ZenoArrow | 6 hours ago
Sure you do, you can prove those that doubt your views wrong.
> Sounds exactly the kind of stuff coding agents thrive at - a verifiable loop. And they can do it 24x7 until done.
Go for it then, you're not putting in any work into it other than giving it a task to do.
dist-epoch | 6 hours ago
ZenoArrow | 5 hours ago
Perhaps the reality is that you know AI needs more hand-holding than this, and the tools aren't up to the task you're thinking of setting them.
dist-epoch | an hour ago
You are also strangely fixated on today's capabilities, completely missing the exponential we are on.
In a few months will have posts here from device driver writers explaining how they hooked up a phone to an Arduino and a video camera and how the AI is automatically writing device drivers.
ZenoArrow | 30 minutes ago
I am talking about today's capabilities because this comment thread started with the suggestion that the benefits of AI for coding was no longer avoidable after the launch of Codex 5.3.
> In a few months will have posts here from device driver writers explaining how they hooked up a phone to an Arduino and a video camera and how the AI is automatically writing device drivers.
A few months? Almost zero chance. If it happens in the next 5 years I'd be less surprised, but I suspect it'll take longer.
fartfeatures | 7 hours ago
ZenoArrow | 7 hours ago
Not really. If AI is just copying someone else's code, it's not really designing it is it. If you want it to truly design something, it needs to be designing it using the same constraints that the human engineers would face, which means it doesn't get the luxury of copying from others, it has to design things like device drivers with the same level of information that human engineers get (e.g. device specifications and information gathered through trial and error).
fartfeatures | 7 hours ago
ZenoArrow | 6 hours ago
No, I'm suggesting in order for it to be a fair test, you need to impose the same restrictions that a human engineer would face.
For example, consider the work done by the Nouveau team in building a set of open source GPU drivers for NVIDIA GPUs. When they started out the specs were not so widely available. They could look at how GPU drivers were developed for other GPUs, but that is not going to be a substitute for exploratory work. Let's see how well AI does at that exploratory work. I think you'll find it's a lot harder than common uses for AI today.
surajrmal | 7 hours ago
dist-epoch | 7 hours ago
erelong | 7 hours ago
AI use should be able to accelerate the development of ports on currently unsupported or undersupported devices which would directly support the project
I guess I wouldn't worry about the policy, they will probably naturally switch it if / when AI becomes more useful in practice
flammafex | 6 hours ago
Fun while it lasted, huh?
ptnpzwqd | 3 hours ago
Coding agents greatly reduce the barrier to contributing something that at least looks okay at the surface, so reviewing contributions will quickly become even more of a bottleneck. Manual contributions used to filter away most low effort attempts, or at least they could easily be identified and rejected.
That dynamic is now different and the maintainers risk being swarmed with low effort contributions, that will take a lot of time to review and respond to. Some AI contributions might be reviewed and revised and overall be of acceptable quality, but how can the maintainers know which without reviewing everything, good and bad alike.
I think we will see multiple attempts like this to shift things back to the old dynamic, by rejecting things that can be identified as AI generated a glance, but I suspect over time it will be difficult to do so, so my prediction is that we will soon see more open source repos stop accepting outside contributions entirely.
Even if LLMs one day will be good enough to quickly produce code that is on par with humans (which I strongly doubt), why would the contributors have any incentive to have someone else do that (the easy part), rather than just doing it themselves?