So what you're saying is somebody took something that was public or at least community-owned and built a fence around it? Or, you might say, an enclosure [1]? Interesting. One could even describe this as rent-seeking behavior [2]?
LLMs themselves are built from the sum of human output when almost nobody who created that got paid for the value they created.
So what I'm hearing is that we have a distortion that can be described in terms of the worker's relationship to the means of production?
I honestly love it when people when people discover socialism from first principles.
That's one way to look at it, another way to look at it would be something like a community farm where someone took a few ingredients that were just sitting there and being given away for free and cooked them up into something special and sold it.
There's nothing wrong with people working on hobby projects for fun, regardless of how useful they may be, and there's nothing wrong with others taking that work and building upon it for their own purposes. That's the entire point of open source software.
If you find yourself upset that someone "stole" your freely given away work or is making a profit off of it or you yourself are not being "compensated" for giving away "free" things you're the problem here.
I think this analogy breaks down when you make money from it. It's more like your chickens lay more eggs than you need, so I break into your coop, steal the excess (you still have all you need) and set up a stall by the side of the road selling eggs for $2 a dozen.
I don't think that follows at all. IF you wanted to use the chicken analogy then they would just be community chickens with freely given away infinite eggs (since it's software) and someone (open source maintainers) because they think it's a lot of fun and are happy to give away their time and labor for free just keep the chickens happy and well-fed.
If you find yourself later being upset that someone took the eggs and made an omelet and sold it and made a few bucks, then you are lying to yourself and others about your reasoning for writing open source code.
If that's the case then you've also been tricked into believing the wrong things about XYZ and separately I can only argue what I'd been tricked to argue because you've stripped away my agency and have accused me of being unable to think for myself.
LLMs are essentially intellectual property laundering as a service. Not how I wanted IP to die, but using this to get more quality content into open source and the good of the commons feels like a critical task for us to share.
I reckon we’re going to see more proprietary software over the next year.
Gone are the days of commercial organisations using OSS as a marketing tactic to gain a base they later monetise. It worked, it was a tactic and no one really liked it.
Your options will be:
a) community led OSS - like Apache projects, CNCF projects, etc
b) some dudes weekend OSS project
c) commercial vendor, likely proprietary or at best ‘source available’
I disagree. My agents are godlike when I ask them to operate software whose previous version's source is in their training data, and pretty confused when I ask them to operate something proprietary. I wouldn't even use proprietary software if the vendor was paying me to do so (unless they were paying me quite a lot) because there's just so much extra friction around it.
The ideal future of software is not as a product or a service but merely as a tool, used by people with a problem to solve that has nothing to do with selling the tool.
This is the first time “open source” has meant anything other than “probably free” to the average person.
I constantly hear people complain about UIs. You know who probably makes a great UI for non-technical retirees? A non-technical retiree. They’ll probably start from forking an OSS project, though.
I think the really clever folks will move up to building vim/emacs/Firefox esque apps that are heavily scriptable, and your weekend OSS projects will be done by non-technical folks who just want something really specific.
> I constantly hear people complain about UIs. You know who probably makes a great UI for non-technical retirees? A non-technical retiree. They’ll probably start from forking an OSS project, though.
You know what makes a great UI for non-technical people? Codex or Claude.
GUIs were a response to the limited resources available on old computers. And, as we're increasingly finding out, text is FINE. People like Teh Pretty(tm), but they'll take plain text if it solves their problem.
But, congratulations, you DID identify the primary bottleneck--customer support. And, no, I am NOT going to be unpaid customer support for non-technical people. Bleah! Get back to me when Claude/Codex don't suck for that.
I work a job for my money that isn't developing and haven't yet developed something worth enough to others to consider making money that way.
If I did, I would consider a finite sum (e.g. kickstart) then release the code. I see software/digital goods as only scarce until they are released. So their value plummets to $0, its only nature defying legal structures that make it otherwise.
I don't think of open source software as a way for me to make money, I think of it as a way for me to prevent other people from making money.
An open source project management software nips at JIRA's profit margins. OSM loosens google maps' stranglehold. The swath of FOSS crms delegitimizes Salesforce's per-seat cost. Most recent example is open weight LLMs driving a spike of fear deep into the heart of the current lynchpin of the American economy.
MIT license was a mistake. We tried to play by the rules of capitalists and the obvious happened: the capitalists exploited us. "Thanks for the free labor! By the way, we don't hire people that don't have FOSS commits on their resume." Google and Facebook, last I checked, forbid the inclusion of GPL3 code in their projects. Great! Let's make more GPL3 copies of their software so nobody has to give them money anymore.
The marxist larping is cringe. In my experience, people that are like this likely have 0 commits to actual OSS.
OSS is a great passion project for a developer that wants to take it on, and when the community comes together, they can do incredible things. From a social perspective, I'd struggle to find anything that's more empowering to the average person than the ability to spin up a Linux server and get going in like 30 minutes.
But support is lacking. Proper documentation can be lacking. The sheer amount of work required to host, fix, update, and keep uptime up on real software at a real business is beyond a hobby, and enterprise support is mandatory, not optional. There are legitimate legal requirements that need to be met.
The fact of the matter is that the nature of the capitalist system that exists in the western nations is the same system that allows for OSS to exist at all. Tailscale was hard as hell to develop. Docker was hard to develop. I love the opensource alternatives to these services. I wouldn't use them at my company.
>I think of it as a way for me to prevent other people from making money
That's a very common misconception I see among the developers. They get upset when someone sells their software, especially when the author distribute it for free.
> That's a very common misconception I see among the developers. They get upset when someone sells their software, especially when the author distribute it for free.
No, that isn't the argument here at all. The argument is that the existence of OSS alternatives (for which there is no real reason to pay more than a nominal distribution fee, aside from contracted support) makes it harder for proprietary software to retain commercial value.
>for which there is no real reason to pay more than a nominal distribution fee, aside from contracted support
FOSS doesn't mean you have to distribute the software for free. If you want to earn money from FOSS, just don't publish it and sell it instead. It doesn't make it less open.
I sell open-source software of myself and others, for example, quite successfully.
Linked article states:
>So if you are redistributing copies of free software, you might as well charge a substantial fee and make some money. Distributing free software is an opportunity to raise funds for development. Don't waste it!
It has been this way all these years, that's why I call it a common misconception. The GNU article by the link above is from 2012 or earlier.
If you thought this through until its final conclusion then you would notice that open source developers would simultaneously be paid more money than proprietary company developers and charge less money than proprietary companies.
People would gladly pay you more because it means they have to pay less.
The only necessary requirement is that the payment goes down with the number of payers and that there is no free rider problem.
I mostly agree, but I don't think that's quite it either. For me, OSS is neither a way for me to make money, nor a way to prevent other people from making money. The latter is closer, but it's not other people making money off of what I create that bothers me, it's that users have to pay someone else in order to use what I created.
I think of OSS as a way to get value into the hands of users and for them to not have to pay someone else for it. I honestly wouldn't be too bothered if they had to pay me, but I relinquish that payment because not paying me means more users and more use, and I have the privilege of a decently paying tech job, so for my personal cost/benefit analysis the free software wins. (And yes, I default to GPL not MIT/BSD.) And it's not just the direct users of my software that I care about, it's also the overall ecosystem of freely available and modifiable software that I want to contribute to.
It also doesn't bother me when other people use OSS as an anti-competition mechanism. That's a different reason to do something that happens to resemble what I'm doing. It doesn't contradict or interfere with what I'm doing nor my reasons for doing it.
Irony: the number of developers who rely upon other people's open source to develop commercial software or services.
Once upon a time, people payed hundreds or thousands of dollars to acquire software development tools. Even those who programmed as a hobby would have to spend a fair bit of money if they wanted to grow beyond interpreted BASIC.
I'm not claiming the current situation is ideal, but open source has given us much more freedom to program.
I'd be more likely to participate in something like that if instead of collecting the money, the maintainer is just requiring that I donate it to the EFF or something like that.
Money as proof of this-guy-actually-cares isn't an awful idea but you gotta be careful to not become a parasite.
Many open-source projects are made for the author's needs, and the author is not interested in implementing feature requests they don't need.
Sending no-strings-attached donations to the person instead of targeted fundraising of the project is just a disservice if you're expecting maintenance, more harm than good.
...and who is going to be motivated to maintain all of those open source apps? You cannot fund all of them, and open source has an unsustainable funding problem.
Even worse, most of open source is now majority slop and it is highly likely that they will be abandoned.
> People call this new way of building “vibecoding.” I hate the name; it makes a real shift in who gets to write software sound like a joke.
Because this is the joke. AI agents didn't change who "gets" to write software at all, it's still the same middle-level and senior devs it's always been. The term is intentionally derogatory towards the people who think otherwise.
Putting misguided effort into positive externalities is certainly a preferable outcome to misguided effort into wasting time (the true limited resource)
Open source developers do not get paid already. I strongly agree developers need to get paid, but you’re not going to stop folks from producing open source for free.
I’d much rather give my money to a small sustainable business or a coop versus a traditional enterprise producing and/or hosting software, whenever possible, especially if they are in some way producing high quality open source software. “Shop local” sort of thing.
Whats wrong with the term vibe coding? If it has a negative connotation, the problem isn't the choice of words, the problem is the software being produced. Rather than change the name, try and figure out how to raise the bar.
I personally think that there's nothing wrong with vibe coding so long as there are as close to watertight ways, checks and balances to verify that the output is correct. And the best way to do that is to invest domain specific expertise in coming up with as comprehensive a test schema as possible.
If the test schema passes and if your original articulation of intent was correct, then what you have is already very good. From that point onwards, it's the question of accelerating development by just reviewing the output of the LLM and getting it into the right shape.
The combination of the review of the output of the LLM and the review of the test schema and its output, that is going to be a lot quicker to do than coding everything from scratch with a proper design paradigm.
> the best way to do that is to invest domain specific expertise
And precisely that is the issue: people vibe coding something and releasing it to the world don't have this domain specific knowledge to build upon. It used to be that building the software would force the programmer to learn about the domain, but that link is broken now too.
Domain experts can also do vibe coding is the principal thesis of what I'm trying to say.
So there's vibe coding for the general populace, and there's vibe coding which domain experts can use to articulate problem statements in well-formed ways and articulate test schemes to ensure that their articulation of the problem is actually solved in the best way possible.
Hopefully that makes sense.
Industry happens because of domain experts for the large part.
If domain experts have a way to accelerate their wherewithal significantly, then industry progresses significantly.
The promise of LLMs and the promise of vibe coding is fundamentally that domain experts of industry can significantly accelerate progress.
The rest, who are not domain experts and who are not of the industry, are also progressing, but the value of the progress they have in terms of how they contribute to how industry contributes to society is fundamentally questionable.
Vibecoding has come to mean "AI slop code" where the people producing it don't understand it, like at all. It can be poorly structured and hard to comprehend.
I believe that train has sailed on this association.
The term was defined to mean slop. If you are vibe coding you are not doing engineering. I assume most customers don't like hearing that the people working on their code aren't aware in any way of the code they are pushing.
I vibereported a few bugs recently. It's satisfying enough to spend 15 minutes on, and easy enough as AI is good at collecting all the details and proofs for you. Also, there's no feeling that you re-used someone else's code through an LLM.
What may or may not help you is the fact that you can in fact just ban these people.
On first sight. And delete the bug report.
They're not bound by any rules of "civility" and neither are you.
It's still eye-roll-inducing, of course, but it's n+1 clicks on GitHub, where n is the number of bogus issues reported (+ their delete buttons) and the +1 is the "block user" button.
Without open source your average AI probably couldn't even write "hello world". We're going to get slop after slop if developer don't write good open source code anymore.
Large software collaborations between strangers on the internet are becoming impossible, because people just send in slop nowadays. It used to be that if someone had written a patch, they had spent some time thinking about the problem and had some level of understanding of what they were doing. Now people ask an LLM to do a contribution to a popular project and just send in whatever the bot produces. Free software is drowning in slop.
In its current form, it's a fundamentally antisocial technology (or at the very least is massively enabling antisocial behaviours). And there's no interest from its makers in fixing that. Hell, if you're a bit cynical you could say its working as intended.
Software is like food. Up until now most people had little choice but to buy meals from restaurants. Programmers are like chefs. Most of them work as line chefs at restaurants. Some of them start their own restaurants. Some cook for friends or volunteer at soup kitchens. Now everyone can have a personal robot chef for $20/mo that acts like C3PO. I think we're going to have to rethink a lot of assumptions about what most people are going to choose when it comes to software now.
That is definitely the case. I think what is clear now is that we're going to have a lot more in-house solutions. Cooking at home became easier. So that the restaurants have to be quite good at what they are doing, and still not every person will turn to them.
But the robot chef is $20/mo on a special deal may rise to $400/mo after the trial period, is more resource-intensive than a human chef, and is one misconfigured guardrail away from putting two tablespoons of nutmeg in your dinner.
...and there is still a place for the robot chef, but the ideal destination for this world of post-commodified-code is probably a sandwich or Stouffer's Instant Dinner: Cheap, reliable, flexible products that don't promise to be a human.
The robot chef routinely poisons people. When it happens it's considered to be a "user" issue or something that will be fixed one day soon. "It's as bad at poisoning people as it will ever be, it'll only ever get better" some people argue.
In the meantime you are encouraged to eat the food it cooks but to take full responsibility if it poisons you or any of your guests.
Normal chefs sometimes poison you too but the most popular restaurants have thousands of happy unpoisoned diners who have all eaten the same dishes. For free.
> Pick one open source tool you use. Find the most annoying thing about installing or running it. Spend a weekend fixing it, with the people who maintain it. Then tell me about it.
My open source projects were barely sustainable when the few PRs that came in would fix actual, non-hypothetical, serious logic problems.
Now I've been seeing PRs like that, fixing annoyances, performance edge cases, or hypotheticals, across my open source repositories, and I don't have time for them. I especially don't have time to fix problems that probably nobody will ever experience, and yet they keep coming. I have not organized my life around triaging annoyances (let alone fixing them).
If the trend continues, I'll soon be forced to close repositories to contributions, or just stop maintaining the projects entirely. And I imagine that's the point a lot of maintainers are at.
> to fix problems that probably nobody will ever experience
Have you considered that the person submitting the PR to fix said problem has themselves experienced it? I find that is most often the case. But there is definitely an increase in the "performance edge cases, or hypotheticals" framing, where I am equally befuddled/too tired.
not the parent, but I received one PR adding a new feature to my and similiar projects at the same time.
When I questioned the use case the PR sat around until it was closed months later.
Hardly a real user.
Before AI,
I have submit PR band-aids bugs I personally experienced, but don't bother enough to fix it properly.
When the maintainer ask about it, I move on and ignore it. I just don't care. Some maintainer found them useful, some don't.
I came from the "free software as in free puppy" camp. I would stop by to play with the puppy, but I won't feed it.
I guess if I do the same today, I might submit a better-than-band-aid fix from LLM. You would probably think I am not a real user.
Sounds more like a real user than an LLM, which would have given you a 4-heading treatise where the third heading was some variation on “Why this matters”.
> Have you considered that the person submitting the PR to fix said problem has themselves experienced it?
I had multiple PRs for entirely hallucinated problems.
Or where no human ever looked at project or PR and were just attacking open source projects with spam PRs pretending to be contributions. Not sure whether it was out of malice or done for some other reason. Maybe they were mistakenly thinking it is helpful?
> Not sure whether it was out of malice or done for some other reason. Maybe they were mistakenly thinking it is helpful?
I believe a fair bit of this is an attempt to gain numbers where quantifiable numbers are advertised on otherwise indistinguishable resumes (with the corresponding belief that those numbers make differences in filtering or hiring decisions).
I base that on having seen people in the days of old put their Stack Overflow reputation on resumes thinking that having a rep of 4321 will be more likely hired than someone with a rep of 1234.
Having "210 accepted PRs on GitHub" (they believe) would appear to make them a more productive / less risky hire than someone who hasn't done any.
---
This is really a case of indistinguishable resumes where nearly everyone in the pool went to the same college, same internships, same capstone project, and maybe the same first year or two of professional experience.
There are job applications that list having a certain amount of contributions to github projects as a requirement. Encourages useless spam. Also funny the implication that only projects on github have any worth.
That wouldn't surprise me in the slightest. I see it with applicants for a position that doesn't have GitHub contributions listed as either a requirement or a nice to have.
I could see it as a "proof of familiarity with CI and expectations for a PR", though for applicants that are spamming the process with poor PRs (and people posting poor Q&A on Stack Overflow back in the day), this often hurts them more than it helps as its a demonstration of what they are going to be doing if they are hired (I've already got too many poor quality PRs to review, your PRs would not reduce the amount of time that the reviewer would need to spend).
That's not the case all the time anymore. It used to be. But it's not now.
So yeah, I've considered it and eliminated it for most PRs we used to get. Because no, a 2nd year student in university did not run into an obscure edge case in production that none of my 1M+ QPS users did.
Sure, but maybe the reason you're getting PRs is because you have 1M+ "QPS users", whereas I can't even think off the top of my head what QPS stands for.
And personally I have never once opened a PR for any issue I didn't personally experience. Even when I used an LLM to assist, I've only worked on things I personally needed.
Actually IMHO, we need better automatic build system integration with feature management. How cool would it be if the average user could vote with their feet if something is needed. Once there is enough adoption and no negative test reports consider merging it.
I guess also AI managed canary l releases could be an option IMHO that combine open PRs.
Just make a contributor license agreement based on Benedictine doctrine. At least one very successful FOSS project does that: https://sqlite.org/codeofethics.html
AI can help with that! Give an agent some rules, like via a skill for the project, and then point it at the PR/Issue for triage to determine if it's one that fits the rules for auto-closing (already reported and decided WONTFIX, for example). Have it run this across all your repos.
Have the AI tooling do the heavy lifting of reproduction and benchmarking. One of my projects is fairly complicated and one user ran into some problems which he could describe the situation that caused it, but was light on any other details: my apt package cache was slow to serve the PHP update to his users. I asked AI to try to reproduce it: Set up a repo, set up different versions of my cache, have a client pull it, put an updated package in there, and have a client pull it again under different configurations. This could have taken me hours to do myself. AI did it with ~2 minutes of my effort.
>I have not organized my life around triaging annoyances
Then don't. AI can help with the triaging. Or, I give you permission: Close the repo for contributions. If you don't have time to maintain it, let people know, so they can fork or replace it. It sounds like your repos are already closed to contributions, due to lack of time/attention. That's fine, don't feel bad, just go ahead and make it official.
> AI can help with that! Give an agent some rules, like via a skill for the project, and then point it at the PR/Issue for triage to determine if it's one that fits the rules for auto-closing (already reported and decided WONTFIX, for example). Have it run this across all your repos.
For me it was a bit useful but it has not offset even 10% of damage caused by LLMs
You can say, "go off and fork the project", and you'll get 20 forks, each one person's pet PR applied, and no more future progress without a dedicated maintainer triaging and coordinating all this stuff in one place. You are asking projects that are surviving but not necessarily thriving to simply die. And for what? To satisfy downstream developer demand for their AI-generated contributions to be accepted?
Not all contributions are equally valuable, and that is notwithstanding AI use. AI facilitates volume but not necessarily quality. Why should anyone spend resources benchmarking and reproducing a non-issue, or a PR that will not otherwise pass muster? We can automate a lot, but human judgment is still critical in making things to be used by humans.
>you'll get 20 forks, each one person's pet PR applied
Sure; if the primary repo goes unmaintained, and nobody else steps up to maintain a fork, yes you might get multiple forks that solve a specific users individual issue. I'm not sure there's a way around that.
The comment I was replying to sounded like someone who was unavailable to handle non-AI requests as well. AI contributions is making it worse, though my biggest projects haven't been hit by it yet <knock on wood>.
Low quality contributions is a very real problem, I agree with that.
By running automated AI agents on untrusted inputs (PRs from strangers on a public repo), you open yourself up to attacks. And you could be spending money for somebody else.
>
AI can help with that! Give an agent some rules, like via a skill for the project, and then point it at the PR/Issue for triage to determine if it's one that fits the rules for auto-closing (already reported and decided WONTFIX, for example). Have it run this across all your repos.
This requires a huge setup of the necessary AI infrastructure.
rejecting a PR saying "i don't have the time to deal with this" is 100% valid thing to do. Don't let yourself be taken advantage of, it's your project. If the submitter doesn't like the answer then they an fork and go on their merry way, that's also a 100% valid thing to do and is the backbone of open source.
Forking also doesn't mean completely going your own way! I have multiple repos forked where we and the maintainer disagreed on the direction, so I keep my additions on top of theirs in my own repo just how I like
yeah sadly only 0.0001% of all users and reporters seem to understand this stand, which is oh-so-obvious if you have ever put anything public, but then people use opensource because it was hailed as free, more secure, and even more trustworthy. and as a result everyone and their dog not only seems to blindly prefer it but is then entitled to service which was never promised. what a massive cognitive dissonance indeed, but one that was carefully catered to and fed by the few key opinion leaders in the opensource sphere.
PRs should be rejected by simply not enabling PRs in the first place. If PRs are needed for development, then that should be done in a private repo that is mirrored to a public one. I wouldn't even be surprised if that person'a readme says the project is open to contributions and that pull requests are welcome. I've found this contradiction to be very common, yet it seems like I'm in the minority for believing it's rude.
and how does this get fixed? really - with more management is the only answer. who pays for this time - nobody. then they would blame the maintainer for being slow... and eventually fork if everyone's luck. why not close outright then?
I've had my AI contribute to a bunch of open source so i'm on the other side of this.
For me they're real issues. Half of which i'd found before but never had the time to submit something for, the other half its things the AI will have encountered - and i check to confirm its a real issue.
The AI will follow guidance you leave in a repository. One of them said "no ai", so it reported that to me and I just forked and went on with my life.
There might be people farming for gh contribution i dontk now — but I can assure you the trend will continue, and afaict it is improving software; at least most of my contributions have been merged.
I don't accept PRs anymore for which there isn't an open issue where either maintainers discussed the issue or it has significant discussion among users. I'm trying to use issue labels to indicate if something is ready to accept PRs.
I see a lot of unsolicited PRs (and my case not from actual users but instead university students trying to pad their resume with random open source contributions but while clearly using AI exclusively)
Yeah, I just started auto-closing PRs from new contributors to avoid drive-bys. I want to know that someone actually uses VisiData before they are allowed to propose a change.
Consider taking donations. As a
MIAB user, I love it and would be sad seeing it go away.
That being said, I know if one reaches that point in burnout, money won’t change anything. You are under no obligation to do any of this after all. So in the meantime, thank you. :)
Why not delegate PR approval to a group of trusted contributors? Give 10 people review and merge permissions, require at least two approvals and passing CI checks, and let them handle the day-to-day PRs. Surely distributing the maintenance workload is a better alternative to closing the repository to contributions?
Perhaps it is time for pull requests to come with payment in escrow that gets transferred after the maintainer does a review and approval.
This would eliminate a lot of AI slop, and I wish I had a popular repo so I could try it. I might unfilter and glance at the non-funded PR's once a year.
Isn't that why it says "with the people who maintain it"? Like maybe open an issue or reach out and tell the maintainer, 'hey, I have this issue, and I want to fix it, would you like a PR if I do?' And then if there is interest, maybe a discussion about it as necessary.
This is what I do, and have done, since long before LLMs. And if they aren't interested? If its a big enough problem for me, I maintain my own fork. Forking is the greatest part about open source projects!
That's sad, but don't let yourself get burned out.
Free software and open source licensing, and projects like Debian, are set up so the software survives even the death of the author and the project. They can make their own improvements for their own ecosystem without upstream support. They integrate what comes from upstream whenever they can.
Someone get their contributions accepted by upstream is a nice thing to have, but upstream don't have to accept it, and if they do, it's at their pace.
Secondly, don't let a pile of open issues and PRs bother you, unless they're fundamentally incorrect (in which case let your authorial voice be known). Because it's also useful for users of your software to give each other feedback and help each other out, e.g. "this app doesn't work with the latest macOS changes, but here's the code changes you'd need to fix it" means that users can help each other often for years before the maintainer gets around to looking at issues. If you're on some sort of code forge site, users can organise there. If you're not (and only accept patches by email or whatever), users can organise themselves elsewhere.
… What's your secret? I keep hearing from people on HN about how they're flooded by AI-driven PRs but my own projects get ignored completely. And most of the time that I check out some other random Show HN project, it's not getting much buzz either. For that matter, I commonly see stuff that has hundreds of seemingly organic stars, and flip over to the PRs and it seems pretty dead to me.
Perhaps add some guidance to your README.md or AGENTS.md, saying that you prefer concise issue reports with simple isolated reproduction sets over PRs? And that they must be, as you said, _actual, non-hypothetical_ issues?
And be OK with closing issues that have dubious merit?
With the increasing power and lowering cost of LLMs, I wonder if we should move to a "apply these patches to the latest version for me" model. That way everybody is happy.
So there is a XKCD cartoon that basically says “14 different standards - let’s build one to fix it all … now you have 15 different standards”
Open Source did an end run round that problem by producing
- better quality code, faster, and replacing the arguments over standards with a level playing field marketplace
As we are learning software has eaten everything but everything has not become software literate - there are hospital MRI machines that are only certified for windows XP. Our supermarket fridges think they are cheap IOT devices for teenagers bedrooms.
It’s going to take a decade or more for society to work its way through replacing all its devices (sewages plants are devices too) wit what is a software literate approach - secure, PKI, monitored updated and … open
No fucker will trust a fridge running a blob of binary - a complete SBOM and a repeatable build will be table stakes
And while we could wait for standards what is most likely to happen is each industry wakes up and starts to say “we want a beyond zero based approach, PKI setup this way, the builds are published and run here and publish the firmware”
Walmart might be thinking about this now - but open source will if we are senile be the backbone
And it will finally be funded as the answer is for industry players to fund the fridge-python api project, mostly by hiring the devs. Or the devs go full pococo and get sponsored
I'm a bit worried that we're going to end up with a lot of vibe-coded, incompatible in-house solutions. Giant frameworks and ecosystems will probably still exist, but niche solutions, I'm not sure.
> Pick one open source tool you use. Find the most annoying thing about installing or running it. Spend a weekend fixing it, with the people who maintain it. Then tell me about it.
This reads like Hacktoberfest all over again.
What one person might find "annoying", another maintainer might've purposefully put there. Not all friction is bad, and telling people to go out, remove all friction and then report back to receive a pat on the head is probably going to end poorly.
This reads more like another attack on the commons than someone genuinely invested in FOSS. Which.. uh.. checks out when you check out the author.
__
Please do not take advice on how to interact with FOSS from anyone but the actual people building it.
If you fundamentally disagree, that is fine. You can become the person that builds their own thing and makes their own rules within that space. But only there.
Just because you can build something with the help of AI doesn't mean it even works the way you intended it to. At worst, you probably may not have the ability to actually verify if it even works.
A very common project that most guys who find this "vibe coding" do is make a calorie tracker, so they ask Claude to build them a calorie tracker, then they add some data from an Apple Watch and Garmin. Now they believe they've built a "dieting app."
It's just funny how many of them actually take data from Garmin and an Apple Watch and add a bunch of activities to work out a "calorie burn of the day" and how inaccurate it is.
So far, many FOSS projects seem to be taking a hardline no-LLM policy, since they seem to have determined LLM tooling to be a net negative for their projects. I wonder if that's because of their nature as very large projects, or very public projects.
But I don't think there's a big overlap between the people who could even name one open-source library and the people flooding the market with vibe-coded to-do list apps.
Better open source can leverage better ways to learn and contribute IMHO. I'm working on building blocks for open source projects, using comprehensive code annotations to help developers learn, experiment, fix, and remix. I welcome help, advice, criticism.
I tried many open source software projects before starting these, and I'm a longtime open source developer so I highly value all the work from people working on vi and emacs, Daisy UI and shadcn, Rocket and Loco, and more. These projects are all aiming for public nonprofit social good.
These look really cool and are thought-provoking for me, as I was considering starting a friendly editor and a design system myself ha. You should show HN these.
I think what matters above all else is a commitment to maintain a product. In a world where writing the software is cheap, what I think we're seeing is that people mostly just want to know that it'll be supported.
Lots of people make things now and that's cool and fun. But what people want to see is a track record of support. They want to know that the things they come to use and depend on won't exit the scene just as quickly as they came.
This is just like 'killed by Google' on steroids. Yes you can write way more things things than before, but if you can't sign up maintain them, then they aren't providing value.
Focus is becoming the commodity. Doing one thing really well is what becomes hard when you could be doing 1,000 things (poorly). To be successful, you can't just throw shit at the wall and see what sticks anymore, because nothing is sticking that way. The walls are already covered in shit. You have to have good product taste and a willingness to apply yourself to it for a long time.
> Pick one open source tool you use. Find the most annoying thing about
> installing or running it. Spend a weekend fixing it, with the people who
> maintain it. Then tell me about it.
Flex: The most annoying thing is that it's written in C (and is therefore a pain to cross-compile), and it has a runtime dependency on GNU M4, which depends on gnulib (the GNU anti-portability library). Flex is currently being rewritten to be more complicated and worse by ESR for reasons beyond my comprehension.
I had an LLM write an m4 implementation in Rust based on the POSIX specs, then fuzz-test it against GNU M4 until it was mostly the same (modulo all the regex bugs it found in GNU M4). Then I had an LLM rewrite Flex in Rust, using rust-m4 as an in-process library.
QEMU: It's written in C, has a bunch of huge dependencies, automated test coverage that rounds to zero, their main contribution path is a mailing list that sometimes swallows patches, and the maintainers might ask for a risky patch to be made more risky before they'll accept it. I went through that process once for a few relatively minor fixes and never want to do it again.
I had an LLM write a multi-architecture emulator with a JIT, high test coverage, and a device plugin system that uses WebAssembly.
Busybox: Written in C, huge pain to build. I tried switching to Toybox instead but it's incomplete, the build system involves a pile of shell scripts that require Bash, not any other POSIX-compliant shell (including its own), and the maintainer's response was basically to tell me to fuck off: https://lists.landley.net/pipermail/toybox/2024-October/0095...
I had an LLM write a replacement in Rust that has enough shell builtins to run an autoconf script from within a hermetic build environment, and its only build-time dependency is rustc.
---
You know why LLMs are so great? Because they save me from having to interact with open-source maintainers who intentionally choose to make their software baroque and their contribution process obnoxious.
You want your project to be written only in ANSI C89 and built with Meson and accept patches only via GPG-signed email to a mailing list that silently drops HTML? Fine, have fun, an ASIC somewhere in Iowa is going to spend the next 3 hours letting me forget your project exists.
> The closer reads like people might've made the right call refusing the engagement.
I think there's enough history of my contributions to various open-source projects over the past ~20 years to serve as evidence that I go pretty far out of my way to be accommodating to projects that are open to contributions.
> It also reads like a refusal of true disengagement.
I'm not sure what "refusal of true disengagement" means.
My goal is generally to accomplish something like building a piano recital scoring app, or being able to run an old game on my laptop. There's a few dozen steps needed to achieve that, which incrementally go down the stack until it encounters a library that works well enough that I don't need to debug it. Sometimes this is a very short list -- I've been amazed at how well the Vulkan ecosystem works, ditto the Go stdlib and the major Go ecosystem libraries.
But sometimes it's like falling down a ladder, hitting each rung on the way down, in which case I turn to an LLM to make the ladder go away. A lot of C projects seem to be like this -- it's a conservative choice of language in general, for people who saw the rise of C++/Java/C#/Go/Rust/etc and decided to skip them, so the attitudes and tooling are likewise conservative.
> Maybe you can now work on that, given that your profile says you're retired?
Being retired means that I want to spend more time doing fun things and less time arguing about file descriptor lifetimes on mailing lists.
Yes, sure, but the initial comment is still full of vitriol, and that's why I said what I said.
> I'm not sure what "refusal of true disengagement" means.
It means not being mad anymore at the people and "showing them" by using LLMs to the degree where one feels the need to rant on about their perceived bad decisions on HN.
I _suspect_ you might not be able to fully turn that off, as it sounds like spite might be a driving force. And that's good, but also, you've kinda made it?
So if anything, you could probably be happy that you're retired and that you do not have to have these discussions anymore.
And maybe also see that the people probably did their best. Which may or may not have been very good, but they did so nonetheless.
> but the initial comment is still full of vitriol
If you think this comment is full of vitriol then you should try reading the kinds of emails people send to mailing lists.
> It means not being mad anymore at the people and "showing them" by
> using LLMs to the degree where one feels the need to rant on about
> their perceived bad decisions on HN.
I think you're reading things into my comment that aren't there.
HN is full of threads lamenting that projects are seeing a decline in community interaction and contributions because LLMs let people vibe-code whatever software they want. I hesitate to use the word "universal", but there's definitely a strong feeling that greenfield LLM-generated projects are in some sense stealing effort or attention that could be put toward an existing project.
I'm sympathetic to that claim on the surface, but the actual experience is that a lot of projects have only had that sort of interaction in the first place because the maintainers forced it as a pre-requisite to getting a patch accepted.
There's some 30 kLoC library out there that decodes an obscure audio codec, it almost works but you need to add some #ifdefs or whatever so it builds on Clang. The would-be user has to mentally prepare themselves for the next few weeks (or months!) of trying to cajole the maintainer into accepting their patch. And maybe it's never accepted, and now you're stuck, and that project goes into the shelf forever.
The article acts as if people with LLMs are the ones whose behavior needs to change to be more pro-social, but from my perspective its the maintainers' attitude that needs to change (if they want their project to continue receiving attention). There's no reason to put up with a maintainer's dictatorial control over a project any more, because the value of code itself is plummeting to epsilon.
> And maybe also see that the people probably did their best. Which may or
> may not have been very good, but they did so nonetheless.
Outcomes have never been judged by whether someone has done their best, but whether they have done well.
> Outcomes have never been judged by whether someone has done their best, but whether they have done well.
Yes but FOSS is primarily about the journey, actually.
You can (and probably should, if you want to get things done) judge the outcomes and plan accordingly, but it's just people doing stuff in their free time.
That cannot be measured purely with the measuring stick of the general market.
Yes, this also means that a lot of it will be complete garbage, but that's fine and nothing to be upset about. At least beyond general disappointment with the lack of skill and sense of an average human. That's my point.
> less time arguing about file descriptor lifetimes on mailing lists.
Hey, I'm retired too, we can argue about that here!
After working in the open for ~15 years, and generally being enthusiastic about collaborating and making shared common base-level tech internally at companies, I'm finding it shocking how much I don't want to even bother pushing projects to a public repo. Never mind doing any sort of releasing, contributing, "marketing", maintenance, patching, etc.
> I think there's enough history of my contributions to various open-source projects over the past ~20 years to serve as evidence that I go pretty far out of my way to be accommodating to projects that are open to contributions.
I mean, signing email and avoiding HTML sound to me like pretty trivial accommodation. And you only have to set up GPG once; it's not as if all these projects have their own mutually incompatible security/privacy/anonymity schemes.
Your post reminds me about the ongoing trend where people don't spend time together anymore. They prefer to sit in front of their computers and interact online. There's a recent article about the difference in the number of people going out on Friday night in 2025 versus 1975. It's like an 80% drop or something like that. People would rather just sit at home.
People have stopped having kids as well -- because it's fundamentally a very human interaction where you have to deal with a lot of bullshit, not just with the kids, but with all the teachers, schools, daycare providers and coaches that you'll end up dealing with. People just opted out. Instagram's more fun.
Software brought people together because it's a team sport and it's hard to ship anything significant on your own, especially in a short time frame. But AI has changed all that. Now we can bring all the isolation that we've brought to the rest of society to software development as well.
AI just gives us another way that we don't have to actually work together anymore. And for you, that seems like a big benefit, because it's clear that you didn't enjoy working with the people that ran those projects.
> Your post reminds me about the ongoing trend where people don't spend
> time together anymore. They prefer to sit in front of their computers
> and interact online.
This is, I think, partially a matter of social circles. The kind of people who enjoy the idea of socialization in an abstract sense but don't actually enjoy the process of socializing will spend all their time indoors on their computers. Meanwhile in the real world people are having a blast. Very few people want to spend time typing on web forums when they could be outside shopping, or dancing, or drinking.
> Software brought people together because it's a team sport and it's hard
> to ship anything significant on your own, especially in a short time frame.
> But AI has changed all that. Now we can bring all the isolation that we've
> brought to the rest of society to software development as well.
While trying not to repeat myself unnecessarily, people whose social circles were formed by contributions to online open-source communities have long been viewed by the rest of society as socially impoverished (with some reason). A web forum or chat room isn't real social interaction, not in the same way that going out to a pub or concert or book club is.
LLMs are finally offering us computer nerds a way out. We can have the computer write the code and debug the segfaults and babysit the benchmarks while we go outside and have fun. We are free from the keyboard now.
> people whose social circles were formed by contributions to online open-source communities have long been viewed by the rest of society as socially impoverished
Those communities that are so impoverished were the things that created reusable software that communities could organize around and customize and improve together. One-off replacements are essentially worthless in comparison.
You put up with the people to build things of greater worth.
> is currently being rewritten to be more complicated and worse by ESR
I don't recognize the name "Flex" but I find this claim hard to believe in and of itself, having seen ESR's rants about the complexity of Automake.
> You want your project to be written only in ANSI C89 and built with Meson and accept patches only via GPG-signed email to a mailing list that silently drops HTML? Fine, have fun, an ASIC somewhere in Iowa is going to spend the next 3 hours letting me forget your project exists.
This to me sounds like the obnoxious attitude: not taking a moment to consider why others would care about their language choice, signed email or avoiding distractions like HTML email (support for which could be their "more complicated and worse").
We probably need more apps with fewer features and more bespoke apps for specific domains rather than having businesses cobble together a dozen SaaS offerings.
Rarely do I see such a stupid headline on HN. I know where this comes from and I can sympathize with the author but from his subjective narrow experience he draws the wrong conclusions.
People post a bunch of throwaway microprojects. So what.
There are wide fields, industries, branches of society, communities, groups of friends left without proper apps to do "that thing". Not to do it better, but to do it at all. It is incredible that a much wider amount of people can engage in creating those apps now. That should be encouraged, and not gatekeeped into existing projects that are already drowning in PRs.
> Pick one open source tool you use. [...] Spend a weekend fixing it
Beware: Your best path forward there may be, contrary to part of the thesis in this post, be to vibe code a replacement. There are a couple reasons why:
Expect to encounter significant push-back due to AI contribution. For example, I've had bug fixes to significant problems be rejected because "we are in the future going to be adopting a no-AI policy". Annoying because I spent my time working on it because they did not have a no-AI policy. Amusing to have another contributor contribute the exact same fix, like character for character, under a new PR (referencing my issue, so not done in isolation) and get it accepted.
There may also need to be sweeping changes to the project you are fixing -- which are going to be a burden to the maintainer and likely wouldn't be accepted because of that. A fork may be the way to go here, but also "vibe coding" a replacement might be the best option.
Earlier this year I build a replacement apt cache called apt-cacher-ultra. Part of it was because the existing apt-cachers had reliability issues, but a bigger part was I wanted something that would survive ubuntu.com being under a DDoS for a week or two. apt-cacher-ng was totally useless at that. The path I wanted to try was to have verified known good snapshots of the upstream, background refreshes, pre-fetching of known "hot" packages when the upstream refreshed. HUGE changes.
I've put 100 hours or more into this "vibe coded" alternative (mostly enhancements and fixes for other users use cases). I'm pretty confident that this wouldn't have been something the apt-cacher-ng or apt-cacher maintainers would have been interested in accepting.
We need more apps. Specifically, everyone should have complete ownership over their own software, so either people learn to build it themselves or truly own the tools they use, instead of renting it via subscriptions or losing control over it by the whims and fancies of c-suites, who brick the software which we paid for.
Here's my most-recent real-world problem: I wanted to plug in my midi controller keyboard and just play a piano, so I can practice my choral music. Very basic.
This turned into many hours over several days trying multiple DAWs, trying to get tehm to install, get them to run at all, get them to recognize my controller, get them to play a VST plugin...changing system settings, completely jacking up my sound, restoring it all again.
It was a nightmare.
So instead I fired up Claude, worked with it for about 3 hours, and now I have an *8-channel* VST player that supports all my hardware, and includes a snappy UI, with outstanding performance and sound. It's amazing!
I literally spent less time creating a custom solution than I did trying to get the off-the-shelf stuff to work. I could make my player OSS too, but the effort to do that is huge, for all the reasons this forum would understand.
Bespoke software avoids all of that and allows me to do what I want, with the environment I have. I don't have all the features of an FL Studio, but I have what I need, and that works for me, beautifully, at low cost and effort. What's not to like?
So maybe we should share OSS "templates" instead of full products, projects that are the barebones of a solution that you can clone, and have your LLM modify to suit your needs, on your platform of choice.
For example, perhaps my barebones VST player is mostly good enough for you, but you also want to play sound fonts? Cool, clone my project, instruct your LLM to add sound font support, and add support for your controller or your output device. Or convert it from Gnome to KDE.
The time and cost of doing this is so little now, why not? I'm getting more out of my computer than ever, and I'm happy about that :)
readers here who have never tried this or does not understand what the person is talking bout - dare not speak. really, you have no idea, so please scroll further. those who understand will surely upvote. because _this_ is what savvy and pragmatic people (would and does) do in 2026 in order to get stuff done, perhaps they did otherwise back in 2012, but not anymore.
This is a take that is outdated if not now, then in 3 months.
These has been this ladder of copium that "oh human still has input here for X, oh shoot it doesn't with Claude 4.5, well human is still needed for Y, oh shoot...".
You are suggesting that you can provide context that the model can't on its own. This is a search problem, every iteration it gets better.
We are at the stage in our company where "/goal optimize this" gets better result than any biased human context engineering can provide.
I think it's reasonable that humans need to provide context for this, and situations like this, forever. The GP has a specific problem. The context of the problem exists in the real world, not something that the agent can search for. This isn't a "humans are still important" take. This is custom software for meeting the specific needs and preferences of an individual.
I fully agree with them that the easiest solution is to ask an agent for exactly what you want, and iterate from there. Evaluating and installing and learning off-the-shelf software is often more of a pain than doing it from scratch.
1) If the UX/UI is important for you, the agent can have access to your other work, or to your other conversation and build your tool with that in mind. This is just another search space question. If you have a documented, predictable taste, the models can efficiently produce what you need.
My uni supervisor/professor used to complain about my slop papers, then I started generating them in his writing style by just pulling down all of his papers and blogs, and he doesn't mind it now.
2) In general, the LLMs are great for steering opinion. There is good research on steering political opinion with LLMs. Your taste and opinions can be formed by the LLMs to like the solution that it provides you.
> You are suggesting that you can provide context that the model can't on its own. This is a search problem, every iteration it gets better.
How exactly do you propose that the model should search my thoughts? Of course I can provide context that the model doesn't have. I can provide context that does not exist anywhere on the Internet (unless I got hacked).
> We are at the stage in our company where "/goal optimize this" gets better result than any biased human context engineering can provide.
First, a "this" must exist to optimize. Second, you have to be happy enough with it that optimization is the only problem.
Fully aligned, i echo the above experience strongly.
However, it seems to me like there must be some middle ground, between sharing architecture "ideas" and full on code PRs.
I myself maintain various FOSS projects; and incoming PRs are marginally useful. generally i just close, and copy with claude following my own & project taste (crediting initial contributor).
Yes, right now this is absolutely the best way to build bespoke software, but it is also very wasteful. Think of all the time and the billions of tokens being burned to repeatedly reimplement apps and features. LLMS code fast, but not yet instantaneously or perfectly for complex features.
What we need is more than just OSS templates. We need fully LLM-managed "software libraries". These would be like package managers on steroids. They would be a huge hierarchy of LLM-coded modules. When you ask for an app, your LLM assembles it from the modules that already exist. If you ask for a novel feature, it codes it up and places it into the library.
So, for your VST player example. You would ask for a barebones version with just the features you need, and your LLM would assemble it for you from preexisting library modules. Then you ask for sound font support. If there is an existing sound font module, it would download it and add it into your app. If there wasn't, it would create one and put it into the library, so the next time someone asks, they can quickly get a VST player with sound font support.
I have a similar experience where I’ve been learning the keyboard but didn’t like the available free options and the good paid apps had really expensive subscription plans so I had Claude build me something that runs natively on my iPad, with midi and microphone support and generates custom learning plans for midi pieces that I import.
I’m an engineer but not a software engineer and it’s been pretty interesting trying out vibe coding and seeing the barriers of entry being blown apart.
I’ve been progressing at learning the piano quite a lot faster than I expected with this. I’ve been debating open sourcing it but I’m not sure I want to take on the work of managing it
Oh wow, just did the exact same thing on my end. If I’m struggling with some fundamentals I’ll just have Claude scour the internet for the best instruction and throw it into a lesson
This would work if all LLM contributions were not only open-source, but also public domain or something. Otherwise you'll get a mess of incompatible licences.
>So maybe we should share OSS "templates" instead of full products, projects that are the barebones of a solution that you can clone, and have your LLM modify to suit your needs, on your platform of choice.
as someone who’s been working with MIDI for a hot minute, I’m really puzzled by your experience. either you have a very fussy controller or you received bad advice. most samplers like Kontakt and modellers like OrganTeq (very well regarded organ modeller) literally comes with a standalone version so that you don’t need to fire up a VST engine/DAW.
> I wanted to plug in my midi controller keyboard and just play a piano, so I can practice my choral music. Very basic.
>
> This turned into many hours over several days trying multiple DAWs, trying to get tehm to install, get them to run at all, get them to recognize my controller, get them to play a VST plugin...changing system settings, completely jacking up my sound, restoring it all again.
Hate to say it but this is genuinely just a skill issue.
Resorting to building a DAW tailored to your own lack of understanding because you can't be bothered to learn tools that are extremely widely used and well-documented is the most extreme case of yak-shaving I've ever seen.
Many apps have just been obsoleted by the new tech. I've tried hundreds of food logging apps and written a few. But none are nearly as good as my current daily driver, a Google spreadsheet wrapped by a GrokBot. I just tell it what I ate two or three times per day, voice or text, no attempt to normalize units or food names. It does all of that with nuance and I seldom have to correct anything. Major new features, like deriving a shopping list from the log, cost one short prompt. Or a running commentary by GaryTaubesBot.
This was just a sci-fi dream in 2022. Now I wonder whether apps are collapsing into an inference wrapped singularity.
I'm still hoping we work out a useful taxonomy. Slop coding, vibe coding, AI-assisted coding, perhaps distinguishing coding from design, ...
Hopefully we can identify the useful dimensions and then come up with some terms that fit the important points in the space.
I have a project where I barely look at the code, but I've put a ton of time into iterating over the design and look&feel. I think of it as "high-effort slop coding", which in my mind is meaningfully different from "true vibe coding", but there's too much self-deprecation in the former term for it to be the right one. (It makes sense when referring to your own stuff, but not when applied to others'.)
"Don't do an easy fun thing to make your life better or more enjoyable, do a hard tedious thing purely for other people's benefit for free," basically.
When someone vibes out a new little todo-like app, they are usually thinking of exactly _one_ user: themselves. They get what they want with very low executive effort. That is what's so cool about it!
Good Open Source contrib takes effort, they don't want randos slopping out PRs. Total apples-to-oranges comparison.
The truth is: open-source doesn't come with any support, maintenance or feature request fulfilment by default.
By definition, that's just a code with no strings attached.
Do not expect maintenance or bug/security fixes by default from any project, unless stated otherwise, or there's at least some metrics which hint of maintenance.
We need SOCIAL.md to describe expectations from this project.
Alternative view : for security reasons open source use will shirnk, since it's becoming most dangerous vector of attack.
Future Software Factories will have extra large static harness, that will be have a very few dependencies, e.g, reactjs, bundled into the package. Even nodejs probably will be more of a runtime feature, patched by the cloud.
open source was mostly a feature of not enough software eng resources. Now , since this is in abundance, instead of a shortcut it becomes a risk.
I've pitched the idea on twitter and asked security guys whether someone does something like this:
Say we have 0/1-days in Python 2, which is not maintained. I'm thinking of software + service which will:
- Detect the exploit attempt
- RE and dissect it
- Document it, propose fix
- Publish it somewhere (!)
- Others discover it (!)
- Individual agents apply the fix locally
This way we don't need centralized maintenance, although it will work better with it.
So in this future, we don't use hardened software. We use security through obscurity.
...But also in this future, the source code of all of these programs is stored in context in the cloud of a few tech companies with unknown levels of security around such context?
Considering they likely use the stuff for training, what are the odds its actually air tight for the foreseeable future? The odds are 100% that the contexts will be compromised because that's the conjecture with the open source projects as well, yes?
I think the biggest issue I see is how cutthroat and self-centered the industry is now. At least in the comments it's 100% about what the fastest way to serve yourself is and absolutely no care to share that effort or resources with others.
You don't think it was easier to scratch your own itch instead of host OSS before AI? Of course it was.
Gone are the days of the hippie techno revolution, I guess. It really is "fuck community, I got mine" and its sad to see.
for now it’s hard to make money from a new open source project. people need money, so they can’t work on open source for free. many open source projects that make money today are older projects with an existing community.
I feel like AI has removed bottlenecks to me being able to contribute to open source projects, and the author is spot on with the “don’t be a jerk about it “guidelines”. Understanding the social element, the community function around maintenance of a project is a different ball game every time, and for me has often been its own research project. I remember before AI killing hours figuring out where to start in understanding mame, csound, and other older projects that have had many many different contributors, and weren’t the most welcoming to newbs. Now I have an assistant to ride shotgun with me while I figure out where to contribute, and can focus on whether my contribution is useful. I have this picture of a world of small functioning tribes suddenly being inundated by nomads who wanna join up.
That's not possible. Somethimes, the open source projects take a path you don't want and they don't want. So you have to build yourself the version you want.
If a project ever has a hope of being sold (“Licensed”) commercially, open source is a death knell. In part due to the AI “training”, and in part due to how easy it is for larger actors in the space simply to take the code and do whatever they want with it.
The problem with Open source presently is that it provides no way to sustain the energy used to support it. I have a limited amount of time in my life; I currently spend roughly 1/3rd of it supporting my family. If I use another chunk of time to ‘support’ open source, I’m looking at my work being taken and used for the profit of others without being able to profit from it myself. If I can’t profit from it, I can’t use it as leverage to make it so I don’t need to spend 1/3rd+ of my life working to support my family.
I am currently writing software with the expectation to sell it. I may even give away copies of it, I haven’t decided yet, but the source code won’t be online, and the support documentation will be behind a registration wall, as will the API. Nothing will be visible on the public internet, not for any other reason than AI scrapers and the large companies that use them.
There may be other ways for small software businesses to be sustainable, but at this moment it’s hard to see them.
> I help progressive nonprofits with their tech for a living
This is the primary perspective behind this post, and that's going to look much much different than the devops roles a lot of us have.
> Too many apps need a database, a cache, a background worker, a reverse proxy and a long config file before you see a login screen
These apps need a database/cache/proxy/config because creating a fault-tolerant, performant, flexible system needs these things. The author wants everything to be a single binary running on a $5-a-month server. A lot of these open source tools could be updated to run in this mode, but that's not the configuration that's sustainable when there's a disaster.
You need to look at why these nonprofits are using open source projects in the first place. They probably want something free, and they probably want something with clear documentation. Maybe it's something that they're already used to. I would argue that AI-assisted (or vibe-coded) apps are cheap enough to fill this niche. Documentation is less important if the features of the app are much more narrowly-scoped. If they're already used to another app, then the vibe-coded one could be designed to provide a similar experience.
I've worked with non-profits that have struggled under the weight of the software they're using. Most often the problem is the software being too general in a way that doesn't model the domain well. Cheap custom software would help a great deal.
I build my own apps in order to solve real world problems I have. I don't see myself opening random PRs in OSS repositories only to get most of them rejected or ignored.
I just told DeepSeek to write a running app customized to my own needs (like Strava), and presto I now have a nice app for my running sessions with just the info I want, no ads, no subscriptions, a full map (openstreetmap) of the route, charts, stats, widgets about time, pace, steps, heart rate, calories, elevation, all I need in my own app, just for me (and my wife)
You may not need it, may not like the colors, the UI, the depth, features, etc. That's the beauty of AI it can build whatever you want however you want it, all for less than a dollar in less than an hour
Yes, we need more apps, tailored to every user. From now on I refuse to install any app not customized to my needs
And yes, we should share more, so AI learns from all of us and builds stuff for us with all the shared knowledge of all
> Yes, we need more apps, tailored to every user. From now on I refuse to install any app not customized to my needs
> And yes, we should share more, so AI learns from all of us and builds stuff for us with all the shared knowledge of all
But the output of this is that every running forum is deluged with hundreds of vibe-coded apps that are marginally different from each other - a moderation burden.
Recently I see even mobile apps that are just LLM wrappers (or not even) charging 5 - 20$ for premium versions and can be one shot with claude or codex.
For example, like you mentioned habit tracking, gym tracking, pdf reader, scanners, language learning and so on.
Also, its hard for me to trust their AIs and feed again and again my information to many different apps and their AIs.
If you think about it, you need one claude or codex subscription and you can generate your own apps with your own AI models that I trust more than ones used in the apps.
Recently I started working on open source local app playstore. Would appreciate your feedback and any input - (https://github.com/Ersho/sayn)
The lack of much-improved open source tools is one of the more baffling things for me when it comes to AI. Especially after Sol 6.1 / Opus 5.5 released, you can whip up a fairly solid application in several days. But... where are they?
OSS devs were sinking a mountain of time into relatively unpaid work, doing unglamorous work. Some of it is in backend systems and console utilities that few people will ever appreciate, but a lot of effort went into user-facing things. And yet, so far I found just one program that replaced one of my mainstays - I went with AI-coded SquirrelDisk to replace FileLight. Everything else is the same, and I just don't understand that. Surely not EVERYBODY rolls their own? Surely some OSS people did embrace AI rather than quitting in disgust?
A small example: tagging music files on Linux is horrible. Kid3 is.. it works, but the UX is almost like a parody. Mp3tag runs on windows and it has a macos version and it's a vast improvement. You can run it through bottles but I refuse to believe everybody does that. Surely people that use Linux listen to music, which leads itself to occasionally needing to tag or rename your audio in batches.
I had some spare time so I rolled my own mp3tag clone that would work on Linux, MacOS and Winblows. I added audio playback to it because opening in external program annoyed me. Then I added some functionality for podcasts to it: being able to search for subtitles, playback speed and such. Then I added the ability to convert between formats, using ffmpeg. Then I added a way to measure loudness, some ReplayGain functionality and the like. Then a way to find duplicates, for making sure those random files you have on that old external HDD can be deleted safely, and a way to search through what you have on all disks, even if they're disconnected, so you don't buy some album you already own like a dumbass.
It took me about three weekends of tinkering with it until it was good enough to switch to it. I keep running into small issues but by and large it's been fairly reliable. It's fairly complex, with a lot of buttons in the view, but given that the 'competition' is mp3tag (which is okay but also ugly) or kid3 (which is both ugly and almost unusable) it's just flat out superior, unless you're happy with mp3tag and don't want all the extra complexity. I don't want to offer unpaid tech support to people on GitHub but I'm considering releasing it and just leaving it there with no further communication and maybe somebody will pick it up. If not, it's still better than alternatives even if there are undoubtedly many, many bugs and the codebase is undoubtedly a tangled mess.
Aside from the ugliness of having almost all functionality without any sub-menus that would give Apple designers heart attacks and perhaps more complexity than it should have (I personalized it for what I wanted, after all), it's genuinely a decent application. It's written in c++, a language I never used. It's fairly performant on my ~15k audio files library. Changing it is as simple as pointing a modern LLM at it and asking in english for some changes. It took me approximately 20 hours to make and the cost was about $10 in subsidized tokens from Claude.
If I, a random guy who doesn't even know c++, can write a usable program in three weekends, then where is all the new vibecoded software? There are so many tools that technically work but the user experience is straight up awful, usually on linux side, or flat out holes that aren't filled, usually on mac side. As best I can tell, very little is being developed. People who have no idea about coding are recompiling console games on pc, in admittedly dubious quality, but people that dedicated years of their lives to OSS... don't?
I'm running into this with just my online friends already. We each have wildly diverse ways of coding now, and our programs don't seem to want to interact with each other. It's like the lisp curse. It's so easy to make your own custom code that you don't bother to even look at what the neighbors are doing.
As AI build-systems become more complex, it might be interesting to publish their work in central locations, and build new standards by which these kinds of software can interact. Or adopt existing ones in new ways!
Right now it's not always clear when and where you can apply an AI workflow and under which rules. This needs to be coordinated sooner or later.
> People call this new way of building “vibecoding.” I hate the name; it makes a real shift in who gets to write software sound like a joke. I’ll call it AI-assisted coding, and I think it could aim somewhere more useful.
AI-assisted means that someone who can actually write code does the piloting of the AI.
Vibecoding however, that's usually people with sometimes zero understanding of tech, often enough not even capable of executing basic CLI commands, "creating" projects, without having anywhere near the skill required to see if they're getting shafted by their AI tool of choice.
No. The old way, where you compromised your workflow because a company, indie dev or maintainer had their own preferences, is dead or dying. Models are now good enough that you can have exactly what you want for little or no extra cost. Most importantly, we don’t have to go through gatekeepers anymore. (OSS maintainers are gatekeepers too: nice, underpaid and overworked, but still gatekeepers.)
My apps are native and fast. They run on all my devices, connect to each other through APIs and MCPs, and look the way I want them to.
I like this, I’ve been trying to do the same. Because of a recent package someone made specifically for you to point an agent at and have it strip your Google Tv of ads, I’ve started adding a bunch of weird apps to the tv. When I realized I could play Mame, I gave it yet another shot. I have made a half dozen attempts at this in my life, both by hand, by buying shitty overseas boxes and a combination without ever finding the arcade experience I want. So I am trying to streamline that for other old folks with little free time who remember arcades fondly by building a curation layer on top of Argosy (which, I think, sits on tom of RoMM) both to limit your choices to a handful of games at once and ensure any new games that you totally acquired legally are what they say they are and will work with your setup.
Because I am a masochist I also decided to make this a way to revive the 2 (!) Stadia controller I have. The initial results were less than promising in the same way hitting your head on bricks is. But with the wireless debugging port wide open on my tv now, I have a small app I am cleaning up that could help you work through the absolute vagaries of connecting Bluetooth game controllers to Google tv (once you go into promiscuous mode, something like three separate services on the tv start fighting for you, last man wins and usually the last man is a service that doesn’t even consider game controllers as a case). I’ve no idea how well it would work or where even to offer it, but in spite of me building new stuff, I mainly do it along these lines too. See also, lintarr a tool I am polishing up which tries to do something like formal verification for all the rules you can set up in different services across the arr stack so you can’t have rules that clash.
ex1fm3ta | 3 hours ago
jmyeet | 3 hours ago
LLMs themselves are built from the sum of human output when almost nobody who created that got paid for the value they created.
So what I'm hearing is that we have a distortion that can be described in terms of the worker's relationship to the means of production?
I honestly love it when people when people discover socialism from first principles.
[1]: https://en.wikipedia.org/wiki/Inclosure_act
[2]: https://en.wikipedia.org/wiki/Rent-seeking
ericmay | 3 hours ago
There's nothing wrong with people working on hobby projects for fun, regardless of how useful they may be, and there's nothing wrong with others taking that work and building upon it for their own purposes. That's the entire point of open source software.
If you find yourself upset that someone "stole" your freely given away work or is making a profit off of it or you yourself are not being "compensated" for giving away "free" things you're the problem here.
pocksuppet | 2 hours ago
ericmay | 2 hours ago
If you find yourself later being upset that someone took the eggs and made an omelet and sold it and made a few bucks, then you are lying to yourself and others about your reasoning for writing open source code.
pocksuppet | 2 hours ago
ericmay | 38 minutes ago
It's kind of a shit argument, why bother with it?
blamestross | 3 hours ago
gwt4life | 3 hours ago
jarym | 3 hours ago
Gone are the days of commercial organisations using OSS as a marketing tactic to gain a base they later monetise. It worked, it was a tactic and no one really liked it.
Your options will be:
a) community led OSS - like Apache projects, CNCF projects, etc
b) some dudes weekend OSS project
c) commercial vendor, likely proprietary or at best ‘source available’
txdv | 2 hours ago
__MatrixMan__ | 2 hours ago
The ideal future of software is not as a product or a service but merely as a tool, used by people with a problem to solve that has nothing to do with selling the tool.
everforward | 2 hours ago
This is the first time “open source” has meant anything other than “probably free” to the average person.
I constantly hear people complain about UIs. You know who probably makes a great UI for non-technical retirees? A non-technical retiree. They’ll probably start from forking an OSS project, though.
I think the really clever folks will move up to building vim/emacs/Firefox esque apps that are heavily scriptable, and your weekend OSS projects will be done by non-technical folks who just want something really specific.
bsder | 2 hours ago
You know what makes a great UI for non-technical people? Codex or Claude.
GUIs were a response to the limited resources available on old computers. And, as we're increasingly finding out, text is FINE. People like Teh Pretty(tm), but they'll take plain text if it solves their problem.
But, congratulations, you DID identify the primary bottleneck--customer support. And, no, I am NOT going to be unpaid customer support for non-technical people. Bleah! Get back to me when Claude/Codex don't suck for that.
pwndByDeath | 3 hours ago
Ampersander | 3 hours ago
phito | 3 hours ago
matkoniecz | 2 hours ago
Even if you limit it to FOSS projects extensively used by massive corporations i is far from rosy.
komali2 | 3 hours ago
An open source project management software nips at JIRA's profit margins. OSM loosens google maps' stranglehold. The swath of FOSS crms delegitimizes Salesforce's per-seat cost. Most recent example is open weight LLMs driving a spike of fear deep into the heart of the current lynchpin of the American economy.
MIT license was a mistake. We tried to play by the rules of capitalists and the obvious happened: the capitalists exploited us. "Thanks for the free labor! By the way, we don't hire people that don't have FOSS commits on their resume." Google and Facebook, last I checked, forbid the inclusion of GPL3 code in their projects. Great! Let's make more GPL3 copies of their software so nobody has to give them money anymore.
Shiggy_ | 2 hours ago
OSS is a great passion project for a developer that wants to take it on, and when the community comes together, they can do incredible things. From a social perspective, I'd struggle to find anything that's more empowering to the average person than the ability to spin up a Linux server and get going in like 30 minutes.
But support is lacking. Proper documentation can be lacking. The sheer amount of work required to host, fix, update, and keep uptime up on real software at a real business is beyond a hobby, and enterprise support is mandatory, not optional. There are legitimate legal requirements that need to be met.
The fact of the matter is that the nature of the capitalist system that exists in the western nations is the same system that allows for OSS to exist at all. Tailscale was hard as hell to develop. Docker was hard to develop. I love the opensource alternatives to these services. I wouldn't use them at my company.
ValdikSS | 2 hours ago
That's a very common misconception I see among the developers. They get upset when someone sells their software, especially when the author distribute it for free.
Please read this: https://www.gnu.org/philosophy/selling.html
If the license doesn't allow to sell free software, the license is not free.
If you don't want others to sell your software, you need to use non-free license. There's absolutely no shame for that.
zahlman | an hour ago
No, that isn't the argument here at all. The argument is that the existence of OSS alternatives (for which there is no real reason to pay more than a nominal distribution fee, aside from contracted support) makes it harder for proprietary software to retain commercial value.
ValdikSS | 32 minutes ago
FOSS doesn't mean you have to distribute the software for free. If you want to earn money from FOSS, just don't publish it and sell it instead. It doesn't make it less open.
I sell open-source software of myself and others, for example, quite successfully.
Linked article states:
>So if you are redistributing copies of free software, you might as well charge a substantial fee and make some money. Distributing free software is an opportunity to raise funds for development. Don't waste it!
It has been this way all these years, that's why I call it a common misconception. The GNU article by the link above is from 2012 or earlier.
imtringued | 2 hours ago
People would gladly pay you more because it means they have to pay less.
The only necessary requirement is that the payment goes down with the number of payers and that there is no free rider problem.
sfink | 51 minutes ago
I think of OSS as a way to get value into the hands of users and for them to not have to pay someone else for it. I honestly wouldn't be too bothered if they had to pay me, but I relinquish that payment because not paying me means more users and more use, and I have the privilege of a decently paying tech job, so for my personal cost/benefit analysis the free software wins. (And yes, I default to GPL not MIT/BSD.) And it's not just the direct users of my software that I care about, it's also the overall ecosystem of freely available and modifiable software that I want to contribute to.
It also doesn't bother me when other people use OSS as an anti-competition mechanism. That's a different reason to do something that happens to resemble what I'm doing. It doesn't contradict or interfere with what I'm doing nor my reasons for doing it.
II2II | 3 hours ago
Once upon a time, people payed hundreds or thousands of dollars to acquire software development tools. Even those who programmed as a hobby would have to spend a fair bit of money if they wanted to grow beyond interpreted BASIC.
I'm not claiming the current situation is ideal, but open source has given us much more freedom to program.
criddell | 2 hours ago
PR coming from an individual? Make the price low but not zero and experiment with the price until the PR rate is manageable.
PR coming from a company? Charge your regular rate.
__MatrixMan__ | 58 minutes ago
Money as proof of this-guy-actually-cares isn't an awful idea but you gotta be careful to not become a parasite.
ValdikSS | 2 hours ago
Sending no-strings-attached donations to the person instead of targeted fundraising of the project is just a disservice if you're expecting maintenance, more harm than good.
rvz | 3 hours ago
Even worse, most of open source is now majority slop and it is highly likely that they will be abandoned.
Xirdus | 3 hours ago
Because this is the joke. AI agents didn't change who "gets" to write software at all, it's still the same middle-level and senior devs it's always been. The term is intentionally derogatory towards the people who think otherwise.
rishankg | 3 hours ago
pwndByDeath | 3 hours ago
pjmlp | 3 hours ago
Rhetoric question of course, especially in the AI coding age.
mmorrell0 | 3 hours ago
toomuchtodo | 3 hours ago
https://rainshare.dev/ enables using up your token subscription to help others with their backlog, for example. Code forge goes brrr.
rvz | 3 hours ago
You cannot eat tokens.
toomuchtodo | 3 hours ago
I’d much rather give my money to a small sustainable business or a coop versus a traditional enterprise producing and/or hosting software, whenever possible, especially if they are in some way producing high quality open source software. “Shop local” sort of thing.
cheney_2004 | 3 hours ago
fork-bomber | 3 hours ago
If the test schema passes and if your original articulation of intent was correct, then what you have is already very good. From that point onwards, it's the question of accelerating development by just reviewing the output of the LLM and getting it into the right shape.
The combination of the review of the output of the LLM and the review of the test schema and its output, that is going to be a lot quicker to do than coding everything from scratch with a proper design paradigm.
pietervdvn | 2 hours ago
And precisely that is the issue: people vibe coding something and releasing it to the world don't have this domain specific knowledge to build upon. It used to be that building the software would force the programmer to learn about the domain, but that link is broken now too.
fork-bomber | 2 hours ago
zahlman | an hour ago
… so, as long as it isn't actually "vibe" coding?
fork-bomber | 38 minutes ago
So there's vibe coding for the general populace, and there's vibe coding which domain experts can use to articulate problem statements in well-formed ways and articulate test schemes to ensure that their articulation of the problem is actually solved in the best way possible.
Hopefully that makes sense.
Industry happens because of domain experts for the large part.
If domain experts have a way to accelerate their wherewithal significantly, then industry progresses significantly.
The promise of LLMs and the promise of vibe coding is fundamentally that domain experts of industry can significantly accelerate progress.
The rest, who are not domain experts and who are not of the industry, are also progressing, but the value of the progress they have in terms of how they contribute to how industry contributes to society is fundamentally questionable.
jmyeet | 3 hours ago
I believe that train has sailed on this association.
rectang | 2 hours ago
zahlman | an hour ago
mtswish | 3 hours ago
NooneAtAll5 | 2 hours ago
why is it bad that customers hear the truth?
Ericson2314 | 3 hours ago
ovnz | 3 hours ago
pwndByDeath | 3 hours ago
pocksuppet | 2 hours ago
hypfer | 2 hours ago
On first sight. And delete the bug report.
They're not bound by any rules of "civility" and neither are you.
It's still eye-roll-inducing, of course, but it's n+1 clicks on GitHub, where n is the number of bogus issues reported (+ their delete buttons) and the +1 is the "block user" button.
sreekanth850 | 3 hours ago
mithril_ | 3 hours ago
Ampersander | 3 hours ago
canelonesdeverd | 3 hours ago
apitman | 3 hours ago
ovnz | 3 hours ago
mipo256 | 3 hours ago
schnevets | 3 hours ago
...and there is still a place for the robot chef, but the ideal destination for this world of post-commodified-code is probably a sandwich or Stouffer's Instant Dinner: Cheap, reliable, flexible products that don't promise to be a human.
pydry | 3 hours ago
In the meantime you are encouraged to eat the food it cooks but to take full responsibility if it poisons you or any of your guests.
Normal chefs sometimes poison you too but the most popular restaurants have thousands of happy unpoisoned diners who have all eaten the same dishes. For free.
barelysapient | 3 hours ago
joshdata | 3 hours ago
My open source projects were barely sustainable when the few PRs that came in would fix actual, non-hypothetical, serious logic problems.
Now I've been seeing PRs like that, fixing annoyances, performance edge cases, or hypotheticals, across my open source repositories, and I don't have time for them. I especially don't have time to fix problems that probably nobody will ever experience, and yet they keep coming. I have not organized my life around triaging annoyances (let alone fixing them).
If the trend continues, I'll soon be forced to close repositories to contributions, or just stop maintaining the projects entirely. And I imagine that's the point a lot of maintainers are at.
cwel | 3 hours ago
Have you considered that the person submitting the PR to fix said problem has themselves experienced it? I find that is most often the case. But there is definitely an increase in the "performance edge cases, or hypotheticals" framing, where I am equally befuddled/too tired.
awakeasleep | 3 hours ago
cwel | 3 hours ago
kuriho | 3 hours ago
j16sdiz | 2 hours ago
I came from the "free software as in free puppy" camp. I would stop by to play with the puppy, but I won't feed it.
I guess if I do the same today, I might submit a better-than-band-aid fix from LLM. You would probably think I am not a real user.
stackghost | 2 hours ago
matkoniecz | 3 hours ago
I had multiple PRs for entirely hallucinated problems.
Or where no human ever looked at project or PR and were just attacking open source projects with spam PRs pretending to be contributions. Not sure whether it was out of malice or done for some other reason. Maybe they were mistakenly thinking it is helpful?
shagie | an hour ago
I believe a fair bit of this is an attempt to gain numbers where quantifiable numbers are advertised on otherwise indistinguishable resumes (with the corresponding belief that those numbers make differences in filtering or hiring decisions).
I base that on having seen people in the days of old put their Stack Overflow reputation on resumes thinking that having a rep of 4321 will be more likely hired than someone with a rep of 1234.
Having "210 accepted PRs on GitHub" (they believe) would appear to make them a more productive / less risky hire than someone who hasn't done any.
---
This is really a case of indistinguishable resumes where nearly everyone in the pool went to the same college, same internships, same capstone project, and maybe the same first year or two of professional experience.
JDups | 45 minutes ago
shagie | 40 minutes ago
I could see it as a "proof of familiarity with CI and expectations for a PR", though for applicants that are spamming the process with poor PRs (and people posting poor Q&A on Stack Overflow back in the day), this often hurts them more than it helps as its a demonstration of what they are going to be doing if they are hired (I've already got too many poor quality PRs to review, your PRs would not reduce the amount of time that the reviewer would need to spend).
levkk | 2 hours ago
So yeah, I've considered it and eliminated it for most PRs we used to get. Because no, a 2nd year student in university did not run into an obscure edge case in production that none of my 1M+ QPS users did.
zahlman | an hour ago
Conscat | 2 hours ago
riedel | 2 hours ago
I guess also AI managed canary l releases could be an option IMHO that combine open PRs.
awakeasleep | 3 hours ago
We need the equivalent of that for open source software maintained by people from before 2004.
tombert | 3 hours ago
BarryMilo | 2 hours ago
I rolled my eyes so hard I died.
linsomniac | 3 hours ago
AI can help with that! Give an agent some rules, like via a skill for the project, and then point it at the PR/Issue for triage to determine if it's one that fits the rules for auto-closing (already reported and decided WONTFIX, for example). Have it run this across all your repos.
Have the AI tooling do the heavy lifting of reproduction and benchmarking. One of my projects is fairly complicated and one user ran into some problems which he could describe the situation that caused it, but was light on any other details: my apt package cache was slow to serve the PHP update to his users. I asked AI to try to reproduce it: Set up a repo, set up different versions of my cache, have a client pull it, put an updated package in there, and have a client pull it again under different configurations. This could have taken me hours to do myself. AI did it with ~2 minutes of my effort.
>I have not organized my life around triaging annoyances
Then don't. AI can help with the triaging. Or, I give you permission: Close the repo for contributions. If you don't have time to maintain it, let people know, so they can fork or replace it. It sounds like your repos are already closed to contributions, due to lack of time/attention. That's fine, don't feel bad, just go ahead and make it official.
matkoniecz | 3 hours ago
For me it was a bit useful but it has not offset even 10% of damage caused by LLMs
and is solving least important problem
yeoldeprogramme | 2 hours ago
You are too nice. Anyone who isn’t going to spend time to explain an issue should be automatically rejected.
We’re in this ugly situation where AI causes the problem and people’s idea to fix it is throw in more AI.
Just like throwing technology at climate change when our own obvious habits are destroying the planet. Let’s move to Mars.
flatline | 2 hours ago
Not all contributions are equally valuable, and that is notwithstanding AI use. AI facilitates volume but not necessarily quality. Why should anyone spend resources benchmarking and reproducing a non-issue, or a PR that will not otherwise pass muster? We can automate a lot, but human judgment is still critical in making things to be used by humans.
hardbass | 2 hours ago
zem | an hour ago
linsomniac | 36 minutes ago
Sure; if the primary repo goes unmaintained, and nobody else steps up to maintain a fork, yes you might get multiple forks that solve a specific users individual issue. I'm not sure there's a way around that.
The comment I was replying to sounded like someone who was unavailable to handle non-AI requests as well. AI contributions is making it worse, though my biggest projects haven't been hit by it yet <knock on wood>.
Low quality contributions is a very real problem, I agree with that.
necovek | 2 hours ago
epistasis | 2 hours ago
swiftcoder | 58 minutes ago
linsomniac | 47 minutes ago
I like the idea of leaning on a decision like Jev model to identify such things.
aleph_minus_one | 2 hours ago
This requires a huge setup of the necessary AI infrastructure.
matkoniecz | 3 hours ago
As a maintainer of some projects: at this point I would appreciate far more useful review of already submitted PRs.
As long it is not blind copy-pasting LLM output. That would be highly obnoxious attack.
chasd00 | 3 hours ago
seki285 | 2 hours ago
larodi | 2 hours ago
ravenstine | 2 hours ago
larodi | 2 hours ago
athrowaway3z | 2 hours ago
For me they're real issues. Half of which i'd found before but never had the time to submit something for, the other half its things the AI will have encountered - and i check to confirm its a real issue.
The AI will follow guidance you leave in a repository. One of them said "no ai", so it reported that to me and I just forked and went on with my life.
There might be people farming for gh contribution i dontk now — but I can assure you the trend will continue, and afaict it is improving software; at least most of my contributions have been merged.
verst | 2 hours ago
I see a lot of unsolicited PRs (and my case not from actual users but instead university students trying to pad their resume with random open source contributions but while clearly using AI exclusively)
saulpw | 2 hours ago
groundzeros2015 | 2 hours ago
no_time | 2 hours ago
That being said, I know if one reaches that point in burnout, money won’t change anything. You are under no obligation to do any of this after all. So in the meantime, thank you. :)
merth | 2 hours ago
dwoldrich | 2 hours ago
This would eliminate a lot of AI slop, and I wish I had a popular repo so I could try it. I might unfilter and glance at the non-funded PR's once a year.
limagnolia | an hour ago
This is what I do, and have done, since long before LLMs. And if they aren't interested? If its a big enough problem for me, I maintain my own fork. Forking is the greatest part about open source projects!
amiga386 | an hour ago
Free software and open source licensing, and projects like Debian, are set up so the software survives even the death of the author and the project. They can make their own improvements for their own ecosystem without upstream support. They integrate what comes from upstream whenever they can.
Someone get their contributions accepted by upstream is a nice thing to have, but upstream don't have to accept it, and if they do, it's at their pace.
Secondly, don't let a pile of open issues and PRs bother you, unless they're fundamentally incorrect (in which case let your authorial voice be known). Because it's also useful for users of your software to give each other feedback and help each other out, e.g. "this app doesn't work with the latest macOS changes, but here's the code changes you'd need to fix it" means that users can help each other often for years before the maintainer gets around to looking at issues. If you're on some sort of code forge site, users can organise there. If you're not (and only accept patches by email or whatever), users can organise themselves elsewhere.
zahlman | an hour ago
mceachen | an hour ago
And be OK with closing issues that have dubious merit?
jjsullivan5196 | an hour ago
> Let me be the agent of a trove of people telling you what to do and how to do it
Do you see the problem here?
exe34 | 8 minutes ago
lifeisstillgood | 3 hours ago
Open Source did an end run round that problem by producing
- better quality code, faster, and replacing the arguments over standards with a level playing field marketplace
As we are learning software has eaten everything but everything has not become software literate - there are hospital MRI machines that are only certified for windows XP. Our supermarket fridges think they are cheap IOT devices for teenagers bedrooms.
It’s going to take a decade or more for society to work its way through replacing all its devices (sewages plants are devices too) wit what is a software literate approach - secure, PKI, monitored updated and … open
No fucker will trust a fridge running a blob of binary - a complete SBOM and a repeatable build will be table stakes
And while we could wait for standards what is most likely to happen is each industry wakes up and starts to say “we want a beyond zero based approach, PKI setup this way, the builds are published and run here and publish the firmware”
Walmart might be thinking about this now - but open source will if we are senile be the backbone
And it will finally be funded as the answer is for industry players to fund the fridge-python api project, mostly by hiring the devs. Or the devs go full pococo and get sponsored
mipo256 | 3 hours ago
ovnz | 3 hours ago
hypfer | 3 hours ago
This reads like Hacktoberfest all over again.
What one person might find "annoying", another maintainer might've purposefully put there. Not all friction is bad, and telling people to go out, remove all friction and then report back to receive a pat on the head is probably going to end poorly.
This reads more like another attack on the commons than someone genuinely invested in FOSS. Which.. uh.. checks out when you check out the author.
__
Please do not take advice on how to interact with FOSS from anyone but the actual people building it.
If you fundamentally disagree, that is fine. You can become the person that builds their own thing and makes their own rules within that space. But only there.
apitman | 3 hours ago
faangguyindia | 3 hours ago
A very common project that most guys who find this "vibe coding" do is make a calorie tracker, so they ask Claude to build them a calorie tracker, then they add some data from an Apple Watch and Garmin. Now they believe they've built a "dieting app."
It's just funny how many of them actually take data from Garmin and an Apple Watch and add a bunch of activities to work out a "calorie burn of the day" and how inaccurate it is.
Apple and Garmin are not accurate for this, read this: https://macrocodex.app/knowledge/rethink/how-many-calories-d...
But very few of them actually bother to check whether their "calorie burn estimate" even works to achieve the rate of loss or gain they set.
Not everyone is "doing better with AI", many of them are doing worse, they just don't know!
komali2 | 3 hours ago
florieger | 3 hours ago
[OP] pandemicsoul | 3 hours ago
FinnLobsien | 3 hours ago
But I don't think there's a big overlap between the people who could even name one open-source library and the people flooding the market with vibe-coded to-do list apps.
jph | 3 hours ago
1. Friendly editor: Vix @ https://github.com/vixide
2. Broader web UI: Lily @ http://github.com/lilydesignsystem
3. Smarter web service helpers: Axa @ https://github.com/publicprojects-io/axa
I tried many open source software projects before starting these, and I'm a longtime open source developer so I highly value all the work from people working on vi and emacs, Daisy UI and shadcn, Rocket and Loco, and more. These projects are all aiming for public nonprofit social good.
listenfaster | 2 hours ago
deepsquirrelnet | 3 hours ago
Lots of people make things now and that's cool and fun. But what people want to see is a track record of support. They want to know that the things they come to use and depend on won't exit the scene just as quickly as they came.
This is just like 'killed by Google' on steroids. Yes you can write way more things things than before, but if you can't sign up maintain them, then they aren't providing value.
Focus is becoming the commodity. Doing one thing really well is what becomes hard when you could be doing 1,000 things (poorly). To be successful, you can't just throw shit at the wall and see what sticks anymore, because nothing is sticking that way. The walls are already covered in shit. You have to have good product taste and a willingness to apply yourself to it for a long time.
x3haloed | 3 hours ago
The fact that free software is benefitting at all from more coding power is a win. We have bottlenecks to solve though.
jmillikin | 2 hours ago
I had an LLM write an m4 implementation in Rust based on the POSIX specs, then fuzz-test it against GNU M4 until it was mostly the same (modulo all the regex bugs it found in GNU M4). Then I had an LLM rewrite Flex in Rust, using rust-m4 as an in-process library.
QEMU: It's written in C, has a bunch of huge dependencies, automated test coverage that rounds to zero, their main contribution path is a mailing list that sometimes swallows patches, and the maintainers might ask for a risky patch to be made more risky before they'll accept it. I went through that process once for a few relatively minor fixes and never want to do it again.
I had an LLM write a multi-architecture emulator with a JIT, high test coverage, and a device plugin system that uses WebAssembly.
Busybox: Written in C, huge pain to build. I tried switching to Toybox instead but it's incomplete, the build system involves a pile of shell scripts that require Bash, not any other POSIX-compliant shell (including its own), and the maintainer's response was basically to tell me to fuck off: https://lists.landley.net/pipermail/toybox/2024-October/0095...
I had an LLM write a replacement in Rust that has enough shell builtins to run an autoconf script from within a hermetic build environment, and its only build-time dependency is rustc.
---
You know why LLMs are so great? Because they save me from having to interact with open-source maintainers who intentionally choose to make their software baroque and their contribution process obnoxious.
You want your project to be written only in ANSI C89 and built with Meson and accept patches only via GPG-signed email to a mailing list that silently drops HTML? Fine, have fun, an ASIC somewhere in Iowa is going to spend the next 3 hours letting me forget your project exists.
hypfer | 2 hours ago
It also reads like a refusal of true disengagement. Maybe you can now work on that, given that your profile says you're retired?
__
That said, I use AI the same way. It saves one from a ton of draining interactions, ego management etc.
jmillikin | 2 hours ago
I think there's enough history of my contributions to various open-source projects over the past ~20 years to serve as evidence that I go pretty far out of my way to be accommodating to projects that are open to contributions.
> It also reads like a refusal of true disengagement.
I'm not sure what "refusal of true disengagement" means.
My goal is generally to accomplish something like building a piano recital scoring app, or being able to run an old game on my laptop. There's a few dozen steps needed to achieve that, which incrementally go down the stack until it encounters a library that works well enough that I don't need to debug it. Sometimes this is a very short list -- I've been amazed at how well the Vulkan ecosystem works, ditto the Go stdlib and the major Go ecosystem libraries.
But sometimes it's like falling down a ladder, hitting each rung on the way down, in which case I turn to an LLM to make the ladder go away. A lot of C projects seem to be like this -- it's a conservative choice of language in general, for people who saw the rise of C++/Java/C#/Go/Rust/etc and decided to skip them, so the attitudes and tooling are likewise conservative.
> Maybe you can now work on that, given that your profile says you're retired?
Being retired means that I want to spend more time doing fun things and less time arguing about file descriptor lifetimes on mailing lists.
hypfer | 2 hours ago
> I'm not sure what "refusal of true disengagement" means.
It means not being mad anymore at the people and "showing them" by using LLMs to the degree where one feels the need to rant on about their perceived bad decisions on HN.
I _suspect_ you might not be able to fully turn that off, as it sounds like spite might be a driving force. And that's good, but also, you've kinda made it? So if anything, you could probably be happy that you're retired and that you do not have to have these discussions anymore.
And maybe also see that the people probably did their best. Which may or may not have been very good, but they did so nonetheless.
jmillikin | 2 hours ago
HN is full of threads lamenting that projects are seeing a decline in community interaction and contributions because LLMs let people vibe-code whatever software they want. I hesitate to use the word "universal", but there's definitely a strong feeling that greenfield LLM-generated projects are in some sense stealing effort or attention that could be put toward an existing project.
I'm sympathetic to that claim on the surface, but the actual experience is that a lot of projects have only had that sort of interaction in the first place because the maintainers forced it as a pre-requisite to getting a patch accepted.
There's some 30 kLoC library out there that decodes an obscure audio codec, it almost works but you need to add some #ifdefs or whatever so it builds on Clang. The would-be user has to mentally prepare themselves for the next few weeks (or months!) of trying to cajole the maintainer into accepting their patch. And maybe it's never accepted, and now you're stuck, and that project goes into the shelf forever.
The article acts as if people with LLMs are the ones whose behavior needs to change to be more pro-social, but from my perspective its the maintainers' attitude that needs to change (if they want their project to continue receiving attention). There's no reason to put up with a maintainer's dictatorial control over a project any more, because the value of code itself is plummeting to epsilon.
Outcomes have never been judged by whether someone has done their best, but whether they have done well.hypfer | 2 hours ago
Yes but FOSS is primarily about the journey, actually.
You can (and probably should, if you want to get things done) judge the outcomes and plan accordingly, but it's just people doing stuff in their free time.
That cannot be measured purely with the measuring stick of the general market.
Yes, this also means that a lot of it will be complete garbage, but that's fine and nothing to be upset about. At least beyond general disappointment with the lack of skill and sense of an average human. That's my point.
s3graham | an hour ago
Hey, I'm retired too, we can argue about that here!
After working in the open for ~15 years, and generally being enthusiastic about collaborating and making shared common base-level tech internally at companies, I'm finding it shocking how much I don't want to even bother pushing projects to a public repo. Never mind doing any sort of releasing, contributing, "marketing", maintenance, patching, etc.
zahlman | an hour ago
I mean, signing email and avoiding HTML sound to me like pretty trivial accommodation. And you only have to set up GPG once; it's not as if all these projects have their own mutually incompatible security/privacy/anonymity schemes.
rpdillon | 2 hours ago
People have stopped having kids as well -- because it's fundamentally a very human interaction where you have to deal with a lot of bullshit, not just with the kids, but with all the teachers, schools, daycare providers and coaches that you'll end up dealing with. People just opted out. Instagram's more fun.
Software brought people together because it's a team sport and it's hard to ship anything significant on your own, especially in a short time frame. But AI has changed all that. Now we can bring all the isolation that we've brought to the rest of society to software development as well.
AI just gives us another way that we don't have to actually work together anymore. And for you, that seems like a big benefit, because it's clear that you didn't enjoy working with the people that ran those projects.
ValdikSS | 2 hours ago
https://news.ycombinator.com/item?id=44810086
jmillikin | 2 hours ago
LLMs are finally offering us computer nerds a way out. We can have the computer write the code and debug the segfaults and babysit the benchmarks while we go outside and have fun. We are free from the keyboard now.
rpdillon | an hour ago
Those communities that are so impoverished were the things that created reusable software that communities could organize around and customize and improve together. One-off replacements are essentially worthless in comparison.
You put up with the people to build things of greater worth.
zahlman | an hour ago
I don't recognize the name "Flex" but I find this claim hard to believe in and of itself, having seen ESR's rants about the complexity of Automake.
> You want your project to be written only in ANSI C89 and built with Meson and accept patches only via GPG-signed email to a mailing list that silently drops HTML? Fine, have fun, an ASIC somewhere in Iowa is going to spend the next 3 hours letting me forget your project exists.
This to me sounds like the obnoxious attitude: not taking a moment to consider why others would care about their language choice, signed email or avoiding distractions like HTML email (support for which could be their "more complicated and worse").
cush | 2 hours ago
poisonborz | 2 hours ago
People post a bunch of throwaway microprojects. So what.
There are wide fields, industries, branches of society, communities, groups of friends left without proper apps to do "that thing". Not to do it better, but to do it at all. It is incredible that a much wider amount of people can engage in creating those apps now. That should be encouraged, and not gatekeeped into existing projects that are already drowning in PRs.
dario_amodei | 2 hours ago
junixx | 2 hours ago
- Old proverb
linsomniac | 2 hours ago
Beware: Your best path forward there may be, contrary to part of the thesis in this post, be to vibe code a replacement. There are a couple reasons why:
Expect to encounter significant push-back due to AI contribution. For example, I've had bug fixes to significant problems be rejected because "we are in the future going to be adopting a no-AI policy". Annoying because I spent my time working on it because they did not have a no-AI policy. Amusing to have another contributor contribute the exact same fix, like character for character, under a new PR (referencing my issue, so not done in isolation) and get it accepted.
There may also need to be sweeping changes to the project you are fixing -- which are going to be a burden to the maintainer and likely wouldn't be accepted because of that. A fork may be the way to go here, but also "vibe coding" a replacement might be the best option.
Earlier this year I build a replacement apt cache called apt-cacher-ultra. Part of it was because the existing apt-cachers had reliability issues, but a bigger part was I wanted something that would survive ubuntu.com being under a DDoS for a week or two. apt-cacher-ng was totally useless at that. The path I wanted to try was to have verified known good snapshots of the upstream, background refreshes, pre-fetching of known "hot" packages when the upstream refreshed. HUGE changes.
I've put 100 hours or more into this "vibe coded" alternative (mostly enhancements and fixes for other users use cases). I'm pretty confident that this wouldn't have been something the apt-cacher-ng or apt-cacher maintainers would have been interested in accepting.
_nobody | 2 hours ago
timbert | 2 hours ago
Here's my most-recent real-world problem: I wanted to plug in my midi controller keyboard and just play a piano, so I can practice my choral music. Very basic.
This turned into many hours over several days trying multiple DAWs, trying to get tehm to install, get them to run at all, get them to recognize my controller, get them to play a VST plugin...changing system settings, completely jacking up my sound, restoring it all again.
It was a nightmare.
So instead I fired up Claude, worked with it for about 3 hours, and now I have an *8-channel* VST player that supports all my hardware, and includes a snappy UI, with outstanding performance and sound. It's amazing!
I literally spent less time creating a custom solution than I did trying to get the off-the-shelf stuff to work. I could make my player OSS too, but the effort to do that is huge, for all the reasons this forum would understand.
Bespoke software avoids all of that and allows me to do what I want, with the environment I have. I don't have all the features of an FL Studio, but I have what I need, and that works for me, beautifully, at low cost and effort. What's not to like?
So maybe we should share OSS "templates" instead of full products, projects that are the barebones of a solution that you can clone, and have your LLM modify to suit your needs, on your platform of choice.
For example, perhaps my barebones VST player is mostly good enough for you, but you also want to play sound fonts? Cool, clone my project, instruct your LLM to add sound font support, and add support for your controller or your output device. Or convert it from Gnome to KDE.
The time and cost of doing this is so little now, why not? I'm getting more out of my computer than ever, and I'm happy about that :)
larodi | 2 hours ago
Conscat | 2 hours ago
augment_me | 2 hours ago
These has been this ladder of copium that "oh human still has input here for X, oh shoot it doesn't with Claude 4.5, well human is still needed for Y, oh shoot...".
You are suggesting that you can provide context that the model can't on its own. This is a search problem, every iteration it gets better.
We are at the stage in our company where "/goal optimize this" gets better result than any biased human context engineering can provide.
shmolyneaux | 2 hours ago
I fully agree with them that the easiest solution is to ask an agent for exactly what you want, and iterate from there. Evaluating and installing and learning off-the-shelf software is often more of a pain than doing it from scratch.
augment_me | an hour ago
And the same library is built without the "seed" or baseline?
zahlman | an hour ago
augment_me | 39 minutes ago
1) If the UX/UI is important for you, the agent can have access to your other work, or to your other conversation and build your tool with that in mind. This is just another search space question. If you have a documented, predictable taste, the models can efficiently produce what you need.
My uni supervisor/professor used to complain about my slop papers, then I started generating them in his writing style by just pulling down all of his papers and blogs, and he doesn't mind it now.
2) In general, the LLMs are great for steering opinion. There is good research on steering political opinion with LLMs. Your taste and opinions can be formed by the LLMs to like the solution that it provides you.
zahlman | an hour ago
How exactly do you propose that the model should search my thoughts? Of course I can provide context that the model doesn't have. I can provide context that does not exist anywhere on the Internet (unless I got hacked).
> We are at the stage in our company where "/goal optimize this" gets better result than any biased human context engineering can provide.
First, a "this" must exist to optimize. Second, you have to be happy enough with it that optimization is the only problem.
montenegrohugo | 2 hours ago
However, it seems to me like there must be some middle ground, between sharing architecture "ideas" and full on code PRs.
I myself maintain various FOSS projects; and incoming PRs are marginally useful. generally i just close, and copy with claude following my own & project taste (crediting initial contributor).
nayroclade | 2 hours ago
What we need is more than just OSS templates. We need fully LLM-managed "software libraries". These would be like package managers on steroids. They would be a huge hierarchy of LLM-coded modules. When you ask for an app, your LLM assembles it from the modules that already exist. If you ask for a novel feature, it codes it up and places it into the library.
So, for your VST player example. You would ask for a barebones version with just the features you need, and your LLM would assemble it for you from preexisting library modules. Then you ask for sound font support. If there is an existing sound font module, it would download it and add it into your app. If there wasn't, it would create one and put it into the library, so the next time someone asks, they can quickly get a VST player with sound font support.
jhloa2 | 2 hours ago
I’m an engineer but not a software engineer and it’s been pretty interesting trying out vibe coding and seeing the barriers of entry being blown apart.
I’ve been progressing at learning the piano quite a lot faster than I expected with this. I’ve been debating open sourcing it but I’m not sure I want to take on the work of managing it
compacct27 | 8 minutes ago
red_admiral | an hour ago
ValdikSS | 2 hours ago
I've seen this idea: https://v-it.org/
timbert | an hour ago
rabf | an hour ago
pitchlatte | 26 minutes ago
rewgs | 9 minutes ago
Hate to say it but this is genuinely just a skill issue.
Resorting to building a DAW tailored to your own lack of understanding because you can't be bothered to learn tools that are extremely widely used and well-documented is the most extreme case of yak-shaving I've ever seen.
delichon | 2 hours ago
This was just a sci-fi dream in 2022. Now I wonder whether apps are collapsing into an inference wrapped singularity.
rckclmbr | 2 hours ago
nadis | 2 hours ago
sfink | 27 minutes ago
Hopefully we can identify the useful dimensions and then come up with some terms that fit the important points in the space.
I have a project where I barely look at the code, but I've put a ton of time into iterating over the design and look&feel. I think of it as "high-effort slop coding", which in my mind is meaningfully different from "true vibe coding", but there's too much self-deprecation in the former term for it to be the right one. (It makes sense when referring to your own stuff, but not when applied to others'.)
micimize | 2 hours ago
When someone vibes out a new little todo-like app, they are usually thinking of exactly _one_ user: themselves. They get what they want with very low executive effort. That is what's so cool about it!
Good Open Source contrib takes effort, they don't want randos slopping out PRs. Total apples-to-oranges comparison.
deckar01 | 2 hours ago
ing33k | 2 hours ago
dankobgd | 2 hours ago
ValdikSS | 2 hours ago
https://dri.es/license-only-versus-stewarded-open-source
The truth is: open-source doesn't come with any support, maintenance or feature request fulfilment by default.
By definition, that's just a code with no strings attached.
Do not expect maintenance or bug/security fixes by default from any project, unless stated otherwise, or there's at least some metrics which hint of maintenance.
We need SOCIAL.md to describe expectations from this project.
If that mindset is new to you, read this: https://raccoon.onyxbits.de/blog/bugreport-free-support/
coretx | 2 hours ago
edstarch | 2 hours ago
royraghav28 | 2 hours ago
It's a cycle which breaks without either component missing.
maxdo | 2 hours ago
Future Software Factories will have extra large static harness, that will be have a very few dependencies, e.g, reactjs, bundled into the package. Even nodejs probably will be more of a runtime feature, patched by the cloud.
open source was mostly a feature of not enough software eng resources. Now , since this is in abundance, instead of a shortcut it becomes a risk.
ValdikSS | 2 hours ago
Yes, something like that.
I've pitched the idea on twitter and asked security guys whether someone does something like this:
Say we have 0/1-days in Python 2, which is not maintained. I'm thinking of software + service which will:
This way we don't need centralized maintenance, although it will work better with it.jayd16 | 2 hours ago
...But also in this future, the source code of all of these programs is stored in context in the cloud of a few tech companies with unknown levels of security around such context?
Considering they likely use the stuff for training, what are the odds its actually air tight for the foreseeable future? The odds are 100% that the contexts will be compromised because that's the conjecture with the open source projects as well, yes?
Doesn't that seem concerning?
jayd16 | 2 hours ago
You don't think it was easier to scratch your own itch instead of host OSS before AI? Of course it was.
Gone are the days of the hippie techno revolution, I guess. It really is "fuck community, I got mine" and its sad to see.
jstanley | 2 hours ago
Yes fix bugs and add features etc. for your own personal use, but don't get your hopes up about contributing back, it's usually a waste of time.
gsoff | 2 hours ago
listenfaster | 2 hours ago
meerita | 2 hours ago
gortok | an hour ago
The problem with Open source presently is that it provides no way to sustain the energy used to support it. I have a limited amount of time in my life; I currently spend roughly 1/3rd of it supporting my family. If I use another chunk of time to ‘support’ open source, I’m looking at my work being taken and used for the profit of others without being able to profit from it myself. If I can’t profit from it, I can’t use it as leverage to make it so I don’t need to spend 1/3rd+ of my life working to support my family.
I am currently writing software with the expectation to sell it. I may even give away copies of it, I haven’t decided yet, but the source code won’t be online, and the support documentation will be behind a registration wall, as will the API. Nothing will be visible on the public internet, not for any other reason than AI scrapers and the large companies that use them.
There may be other ways for small software businesses to be sustainable, but at this moment it’s hard to see them.
shmolyneaux | an hour ago
This is the primary perspective behind this post, and that's going to look much much different than the devops roles a lot of us have.
> Too many apps need a database, a cache, a background worker, a reverse proxy and a long config file before you see a login screen
These apps need a database/cache/proxy/config because creating a fault-tolerant, performant, flexible system needs these things. The author wants everything to be a single binary running on a $5-a-month server. A lot of these open source tools could be updated to run in this mode, but that's not the configuration that's sustainable when there's a disaster.
You need to look at why these nonprofits are using open source projects in the first place. They probably want something free, and they probably want something with clear documentation. Maybe it's something that they're already used to. I would argue that AI-assisted (or vibe-coded) apps are cheap enough to fill this niche. Documentation is less important if the features of the app are much more narrowly-scoped. If they're already used to another app, then the vibe-coded one could be designed to provide a similar experience.
I've worked with non-profits that have struggled under the weight of the software they're using. Most often the problem is the software being too general in a way that doesn't model the domain well. Cheap custom software would help a great deal.
kaishin | an hour ago
Kuyawa | an hour ago
You may not need it, may not like the colors, the UI, the depth, features, etc. That's the beauty of AI it can build whatever you want however you want it, all for less than a dollar in less than an hour
Yes, we need more apps, tailored to every user. From now on I refuse to install any app not customized to my needs
And yes, we should share more, so AI learns from all of us and builds stuff for us with all the shared knowledge of all
philipwhiuk | 9 minutes ago
> And yes, we should share more, so AI learns from all of us and builds stuff for us with all the shared knowledge of all
But the output of this is that every running forum is deluged with hundreds of vibe-coded apps that are marginally different from each other - a moderation burden.
Ersho | 59 minutes ago
Also, its hard for me to trust their AIs and feed again and again my information to many different apps and their AIs.
If you think about it, you need one claude or codex subscription and you can generate your own apps with your own AI models that I trust more than ones used in the apps.
Recently I started working on open source local app playstore. Would appreciate your feedback and any input - (https://github.com/Ersho/sayn)
qwoprunner | 48 minutes ago
OSS devs were sinking a mountain of time into relatively unpaid work, doing unglamorous work. Some of it is in backend systems and console utilities that few people will ever appreciate, but a lot of effort went into user-facing things. And yet, so far I found just one program that replaced one of my mainstays - I went with AI-coded SquirrelDisk to replace FileLight. Everything else is the same, and I just don't understand that. Surely not EVERYBODY rolls their own? Surely some OSS people did embrace AI rather than quitting in disgust?
A small example: tagging music files on Linux is horrible. Kid3 is.. it works, but the UX is almost like a parody. Mp3tag runs on windows and it has a macos version and it's a vast improvement. You can run it through bottles but I refuse to believe everybody does that. Surely people that use Linux listen to music, which leads itself to occasionally needing to tag or rename your audio in batches.
I had some spare time so I rolled my own mp3tag clone that would work on Linux, MacOS and Winblows. I added audio playback to it because opening in external program annoyed me. Then I added some functionality for podcasts to it: being able to search for subtitles, playback speed and such. Then I added the ability to convert between formats, using ffmpeg. Then I added a way to measure loudness, some ReplayGain functionality and the like. Then a way to find duplicates, for making sure those random files you have on that old external HDD can be deleted safely, and a way to search through what you have on all disks, even if they're disconnected, so you don't buy some album you already own like a dumbass.
It took me about three weekends of tinkering with it until it was good enough to switch to it. I keep running into small issues but by and large it's been fairly reliable. It's fairly complex, with a lot of buttons in the view, but given that the 'competition' is mp3tag (which is okay but also ugly) or kid3 (which is both ugly and almost unusable) it's just flat out superior, unless you're happy with mp3tag and don't want all the extra complexity. I don't want to offer unpaid tech support to people on GitHub but I'm considering releasing it and just leaving it there with no further communication and maybe somebody will pick it up. If not, it's still better than alternatives even if there are undoubtedly many, many bugs and the codebase is undoubtedly a tangled mess.
Aside from the ugliness of having almost all functionality without any sub-menus that would give Apple designers heart attacks and perhaps more complexity than it should have (I personalized it for what I wanted, after all), it's genuinely a decent application. It's written in c++, a language I never used. It's fairly performant on my ~15k audio files library. Changing it is as simple as pointing a modern LLM at it and asking in english for some changes. It took me approximately 20 hours to make and the cost was about $10 in subsidized tokens from Claude.
If I, a random guy who doesn't even know c++, can write a usable program in three weekends, then where is all the new vibecoded software? There are so many tools that technically work but the user experience is straight up awful, usually on linux side, or flat out holes that aren't filled, usually on mac side. As best I can tell, very little is being developed. People who have no idea about coding are recompiling console games on pc, in admittedly dubious quality, but people that dedicated years of their lives to OSS... don't?
Kim_Bruning | 21 minutes ago
As AI build-systems become more complex, it might be interesting to publish their work in central locations, and build new standards by which these kinds of software can interact. Or adopt existing ones in new ways!
Right now it's not always clear when and where you can apply an AI workflow and under which rules. This needs to be coordinated sooner or later.
mschuster91 | 23 minutes ago
AI-assisted means that someone who can actually write code does the piloting of the AI.
Vibecoding however, that's usually people with sometimes zero understanding of tech, often enough not even capable of executing basic CLI commands, "creating" projects, without having anywhere near the skill required to see if they're getting shafted by their AI tool of choice.
dataminded | 21 minutes ago
My apps are native and fast. They run on all my devices, connect to each other through APIs and MCPs, and look the way I want them to.
tclancy | 18 minutes ago
Because I am a masochist I also decided to make this a way to revive the 2 (!) Stadia controller I have. The initial results were less than promising in the same way hitting your head on bricks is. But with the wireless debugging port wide open on my tv now, I have a small app I am cleaning up that could help you work through the absolute vagaries of connecting Bluetooth game controllers to Google tv (once you go into promiscuous mode, something like three separate services on the tv start fighting for you, last man wins and usually the last man is a service that doesn’t even consider game controllers as a case). I’ve no idea how well it would work or where even to offer it, but in spite of me building new stuff, I mainly do it along these lines too. See also, lintarr a tool I am polishing up which tries to do something like formal verification for all the rules you can set up in different services across the arr stack so you can’t have rules that clash.