Vibe Coding Is Killing Open Source Software, Researchers Argue:
According to a new study from a team of researchers in Europe, vibe coding is killing open-source software (OSS) and it's happening faster than anyone predicted.
Thanks to vibe coding, a colloquialism for the practice of quickly writing code with the assistance of an LLM, anyone with a small amount of technical knowledge can churn out computer code and deploy software, even if they don't fully review or understand all the code they churn out. But there's a hidden cost. Vibe coding relies on vast amounts of open-source software, a trove of libraries, databases, and user knowledge that's been built up over decades.
Open-source projects rely on community support to survive. They're collaborative projects where the people who use them give back, either in time, money, or knowledge, to help maintain the projects. Humans have to come in and fix bugs and maintain libraries.
Vibe coders, according to these researchers, don't give back.
The study Vibe Coding Kills Open Source, takes an economic view of the problem and asks the question: is vibe coding economically sustainable? Can OSS survive when so many of its users are takers and not givers? According to the study, no.
"Our main result is that under traditional OSS business models, where maintainers primarily monetize direct user engagement...higher adoption of vibe coding reduces OSS provision and lowers welfare," the study said. "In the long-run equilibrium, mediated usage erodes the revenue base that sustains OSS, raises the quality threshold for sharing, and reduces the mass of shared packages...the decline can be rapid because the same magnification mechanism that amplifies positive shocks to software demand also amplifies negative shocks to monetizable engagement. In other words, feedback loops that once accelerated growth now accelerate contraction."
[...] According to Koren, vibe-coders simply don't give back to the OSS communities they're taking from. "The convenience of delegating your work to the AI agent is too strong. There are some superstar projects like Openclaw that generate a lot of community interest but I suspect the majority of vibe coders do not keep OSS developers in their minds," he said. "I am guilty of this myself. Initially I limited my vibe coding to languages I can read if not write, like TypeScript. But for my personal projects I also vibe code in Go, and I don't even know what its package manager is called, let alone be familiar with its libraries."
The study said that vibe coding is reducing the cost of software development, but that there are other costs people aren't considering. "The interaction with human users is collapsing faster than development costs are falling," Koren told 404 Media. "The key insight is that vibe coding is very easy to adopt. Even for a small increase in capability, a lot of people would switch. And recent coding models are very capable. AI companies have also begun targeting business users and other knowledge workers, which further eats into the potential 'deep-pocket' user base of OSS."
This won't end well. "Vibe coding is not sustainable without open source," Koren said. "You cannot just freeze the current state of OSS and live off of that. Projects need to be maintained, bugs fixed, security vulnerabilities patched. If OSS collapses, vibe coding will go down with it. I think we have to speak up and act now to stop that from happening."
He said that major AI firms like Anthropic and OpenAI can't continue to free ride on OSS or the whole system will collapse. "We propose a revenue sharing model based on actual usage data," he said. "The details would have to be worked out, but the technology is there to make such a business model feasible for OSS."
[...] "Popular libraries will keep finding sponsors," Koren said. "Smaller, niche projects are more likely to suffer. But many currently successful projects, like Linux, git, TeX, or grep, started out with one person trying to scratch their own itch. If the maintainers of small projects give up, who will produce the next Linux?"
arXiv link: https://arxiv.org/abs/2601.15494
(Score: 5, Touché) by driverless on Monday February 09, @12:35PM (33 children)
How much value is an AI slopper really going to provide to an OSS project? Would anyone actually want them contributing?
(Score: 5, Touché) by turgid on Monday February 09, @12:42PM
The world is soon going to run out of popcorn.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 1) by khallow on Monday February 09, @12:50PM (13 children)
Sounds to me a little like complaining that one isn't getting more money for this additional set of extremely low value customers. I feel it's a bit like someone invents a machine that does the work of 100 people. Ok. So I start a business where I have that one button masher doing the work of 100 people. Should I be paying the button masher 100 times as much? What if the work is worth only a little more than the button masher makes?
Or maybe it's like a car. I "take" the car for a spin (rather than "give" to the car) and put some wear on the tires and engine?
(Score: 3, Touché) by Anonymous Coward on Monday February 09, @02:23PM
More car analogies, please.
(Score: 5, Insightful) by JoeMerchant on Monday February 09, @03:38PM (11 children)
> Should I be paying the button masher 100 times as much?
George Jetson had a 3 day work week, 2 hour workday, just mashing a button, and he could afford a flying car...
🌻🌻🌻🌻 [google.com]
(Score: 3, Touché) by khallow on Monday February 09, @04:40PM (1 child)
(Score: 4, Touché) by JoeMerchant on Monday February 09, @05:08PM
Big Macs weren't a part of the Jetson universe, with Rosie the Robot to do the cooking...
🌻🌻🌻🌻 [google.com]
(Score: 3, Insightful) by turgid on Monday February 09, @04:53PM (8 children)
There was obviously significant wealth redistribution in that society. Currently, we have huge inequality, because we try to force everyone to work as many hours as possible while letting a few hoard as much as possible. In fact, we try to make people who really shouldn't have to, work. We're going backwards.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 5, Touché) by JoeMerchant on Monday February 09, @05:04PM (2 children)
Even the Flintstones were more communist than today - working man knocks off at the whistle, commutes home to a single-family dwelling with a nice yard, his wife, child and pets which he provides for doing rock-digging equipment operator work, plenty of leisure time... Both were written to be "relatable" to suburban Americans of the 1960s. These days we get Squid Game, and the Mad Max series...
🌻🌻🌻🌻 [google.com]
(Score: 4, Touché) by Anonymous Coward on Tuesday February 10, @01:17AM (1 child)
Whereas if the Chinese Government takes some of the wealth and gives it to the people, millions will praise it as Communism. 🤣
The US people don't even want to give free/subsidized healthcare to the poor (even though they already pay for it to be more inefficiently, more expensively and less effectively delivered via ERs 🤣)...
(Score: 5, Insightful) by JoeMerchant on Tuesday February 10, @01:44AM
>The US people don't even want to give free/subsidized healthcare to the poor (even though they already pay for it to be more inefficiently, more expensively and less effectively delivered via ERs 🤣)...
Yeah, there's another thread about the lead leaving the air (and people's hair) around the time it was banned in gasoline... If they nerf the EPA so thoroughly that they start putting lead back into the gasoline, it's time to leave the country.
🌻🌻🌻🌻 [google.com]
(Score: 1) by khallow on Monday February 09, @07:00PM (4 children)
I can see how a total hive like Scotland would have huge inequality. /sarc
My view is that we should think about why labor protections backfire and why so many people have the mindset that "hoarding" - in other words, investment is something that only the "few" do. There's plenty of signs of broken labor and investment systems in the developed world if one chooses to pay attention.
My view is that work is how most people generate wealth in the world and it works really well. But huge parts of the world deliberately interfere with work in feeble attempts to make it better for the worker. What is missed is that this interference decreases the value of that labor to the employer. Employment is a two-way trade. And if you make the supply of labor more expensive then you get less demand.
Similarly, investment is a huge way to accumulate wealth without having to grind for it. When you only view it as a privilege of others, then you're short changing yourself.
(Score: 5, Insightful) by turgid on Monday February 09, @07:40PM (3 children)
What I'm trying to say is that in many of today's societies, including the UK, George Jetson would be working a busy 40-50 hour week for much less money, struggling to pay his mortgage, Mrs Jetson would be working too, similar hours meanwhile that society would be full of the sick, old, disabled and refugees being denied the pittance they need to get by all while being villified by the very wealthy who don't want to help and the not-quite-getting-by workers who are looking for scapegoats.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 3, Touché) by khallow on Tuesday February 10, @12:59PM (2 children)
They aren't so denied, of course. And that "pittance" costs a lot of money. It's a luxury you can afford because your society is wealthy. My take: prioritize the economy that can generate those "pittances" or you won't be able to afford them in the long run.
(Score: 2) by turgid on Tuesday February 10, @06:37PM (1 child)
My take: prioritize the economy that can generate those "pittances" or you won't be able to afford them in the long run.
That's a no-brainer and has been the prevailing economic and political wisdom most places outside of the USSR since forever. The problem is that the next part gets forgotten, overlooked or simply ignored in an effort to ensure that more and more wealth keeps heading upward.
There's plenty of evidence which shows that more equal societies with good (but not total) wealth redistribution and fairer taxes [taxjustice.uk] work better overall.
The greedy like to ignore the fact that some social security is necessary, that society costs money. It means they don't get to hoard as much as they would otherwise. And they work tirelessly [reformparty.uk] to try to convince us that we should vote against our own interests.
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 1) by khallow on Friday February 13, @12:07AM
It's been outside of a lot more than just the USSR. Like outside of most human history.
As to the second paragraph, I already wrote my opinion on the "evidence" - "It's a luxury you can afford because your society is wealthy." This is like peacock feathers or rutting deer. The peacock with the more brilliant feathers or the buck with the best rack and jousting gets the girls. That's because they exhibit fitness. The same goes for societies. Welfare is something that strong, wealthy societies can do. And sure, I can see some value in that. But I also see a number of societies running up large debt bills because they can't really afford the virtue displays - including the UK and the US.
As to fairness? That's a well abused term like "free" or "love". When I see something like (in your link):
I interpret that "take" as "steal". And "super-rich" as "anyone and anything that isn't nailed down". Trickle down theory really works well to describe taxation. The super-rich didn't get that way by failing to avoid taxes. They're pretty mobile and they'll figure a way to escape the tax hammer. The UK is particularly notorious for its history of tax exiles. The lower and middle class doesn't have those advantages or expert help. They'll be the natural next step to cover spending that the "fair" taxes on the wealthy will fail to do.
And I long ago ceased to be impressed by taxes that are applied in themselves as a punitive or society-changing thing rather than as a means to pay for public goods and services that the society needs (welfare can be a need).
(Score: 3, Insightful) by looorg on Monday February 09, @07:31PM
Isn't it that which is the problem? You are creating less and less what we could call "real programmers" and more and more "AI sloppers". The "AI sloppers" won't contribute hardly anything of worth and value, after all they don't really know how the machines work or how to program. They just give instructions to the AI and it slops it together for them. That probably doesn't count as a contribution. If it did someone more competent would slop it together better. Just not enough "real programmers" to go around.
The amount of people that actually learns to code is staggeringly small, and probably shrinking. As the workforce is slowly being filled with more and more sloppers instead of actual coders. After all coding was hard cause we had so little information, now there is to much information which makes it overwhelming and hard plus people just got lazy and now the machines "code" for them.
Most universities or schools hardly teach or put much time into actual real programming anymore and it's probably going to be even less of it in the future. It will be the "it" language for the moment -- for a long time that have been some version of C (C, C++ or C#) and Java. But these days it seems to be more and more Python and such things. One can just watch the TIOBE index. It's all Python, now more then twice as popular and common as C. ASM? Nope. So you kind of get out of it what you put into it. Slop in, Slop out.
(Score: 4, Touché) by c0lo on Monday February 09, @09:31PM (13 children)
How much effort is required to reject AI slop from hitting the main branch of an OSS project?
(no, it's a serious consideration, don't reply with "just use vibe code reviews", it's wouldn't be even funny)
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Insightful) by Bentonite on Tuesday February 10, @12:20AM (2 children)
It would take a lot of effort to reject copyright infringing slop from the master branch of a free software project, because at first glance the commit usually will look plausible - only on closer inspection will it be noticeable that it is random code, that looks right, but is wrong.
(Score: 4, Touché) by c0lo on Tuesday February 10, @12:52AM (1 child)
Oh, the irony for being sued for infringing the copyright of a bad code would be something special.
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by Bentonite on Tuesday February 10, @02:10AM
Even if code is terrible, you can still get sued if you infringe copyright by distributing it in a way that does not follow the license (for example, MIT expat requires including a copy of the license and retaining the copyright information of the copyright holders and copyright year(s)).
Sometimes the copyright holder does nothing, as saying that terrible code was from you would be an embarrassment - for example it is claimed that the developer of the Dirty Operating System copied from some other OS (which was later licensed to Microsoft, who renamed it to MS-DOS - provided the copying happened, the license document would have falsely stated that all copyright was in order).
What's most likely to happen is for a few LLM slop commits to slip through and proceed to cause bugs and only then will the main developers realize that such submitter is maliciously submitting proprietary software and they have to waste their time cleaning out the commits and barring the submitter (the only thing that you can get banned for in a real free software project is intentionally submitting proprietary software).
(Score: 4, Interesting) by JoeMerchant on Tuesday February 10, @02:04AM (9 children)
> "just use vibe code reviews", it's wouldn't be even funny)
It's not funny, but it's part of the answer.
In my recent experience, you can trust the content of a "vibe code review" even less than vibe code, but... it is a worthwhile exercise (I'm going to do one tomorrow morning on a colleague's PR) - the vibe code review says some outright stupid stuff, strike that and move on, then it says some things you actually want to know about - and if you look at what it's pointing at, those are usually correct. Would you have found all that on your own? In the 30 seconds it took to skim through the vibe code review output? That's useful. Do you rely on it 100%? Hell no, but it does produce helpful output that results in a higher quality end product, if you know how to use it.
On the vibe code end of things, how much testing, review, refinement, refactoring needs to be done before it's ready to show to another human? Usually a lot. Are we saving time overall? Depends. In my wheelhouse of the stuff I've done daily for the last 20 years, hell no - I can do that quicker myself, maybe calling on AI for the occasional API call structure I don't know off the top of my head (like reading a paper manual used to suffice for), but overall: when I'm in charge it goes faster. Something weird (to me) like having Rust generate an .svg based website with server sent evnets keeping the display updated in realtime? Uh, yeah, I _could_ do that by hand, but I can direct an AI agent to do (a simple one) about 10x faster than I can look up all that twisted syntax that I haven't spent any significant hands-on time with, ever.
I call AI agents: power tools. Like chain saws. With great power comes great responsibility - if you give the Phoenix AZ High School JV football squad each a chainsaw and tell them to clear 40 trees from around a cabin in Yosemite, without any instruction or oversight? You're going to have some problems with misuse of power tools. Give those same chainsaws to people who know what they're doing and the chainsaws will be very helpful.
AI everything is new, nobody is an expert (and these days I haven't seen any "Wanted: AI programmer with 15 years experience" ads like we used to get for the old new tech of the day). Our company is encouraging us to step up and "share our AI expertise" - I can't imagine that anyone has any. What I, and my colleagues, have are a few months of AI experience - and that's worth sharing, but what worked last November (when I was last doing daily AI programming work) and what works now (I just dove back in yesterday) are somewhat different things - to call it a rapidly evolving field is to underplay the speed with which it is changing, and the changes of the past 6 months applying LLMs to programming tasks have been consistently toward more power, fewer miscues. It's still far from perfect, but the progress is palpable.
🌻🌻🌻🌻 [google.com]
(Score: 4, Insightful) by c0lo on Tuesday February 10, @02:56AM (8 children)
What is the rest?
There's a technique of ensuring a 1/10000 accuracy of a transcription by using transcribers with a 1% accuracy - just give the same task (w/ the same input) to two different/independent transcribers and reconcile the results (the chance of both making an transcription error in the same place is 1%*1%).
I can't quantify how worthwhile the exercise is for the "vibe code review" on a "vibe coded source", the two exercises are not independent.
I was saved by code-linting more times that I like to admit, but it was a deterministic linter (one of the most insidious case is "unused variable", when you want to use the unused object but end up using a different one). Are there cases in which an AI reviewer offers more than a deterministic linter?
While the wood they cut may not be, chain saws are deterministic. Cutting wood of variable quality with an nondeterministic chainsaw is not an an experience I'd dare to acquire, given the risk to life and limb.
"A computer lets you make more mistakes faster than any invention in human history - with the possible exceptions of handguns and tequila" (and chainsaws before the computers)
There have to be early adopters, but I grew old enough to no longer enjoy being among them.
One on top of the other, I'm not against AI in programming, I just don't have an incentive or motivation to try it, my plate is full with things where vibe coding can't help (architecture/integration)
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 1) by khallow on Tuesday February 10, @01:03PM (3 children)
That approach relies on the transcription processes being independent. That's a poor assumption to make for AI. There will be a lot of overlap in algorithms and data sets. They might even be using each others' output as training material.
(Score: 2) by c0lo on Tuesday February 10, @01:58PM (2 children)
Only one para down from the quoted one reads
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 1) by khallow on Tuesday February 10, @07:17PM (1 child)
(Score: 2) by c0lo on Tuesday February 10, @09:31PM
I'm not proposing anything, I set a context to contrast with the "vibe review a vibed code", to support why I can't trust the resulted code as being better.
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Interesting) by JoeMerchant on Tuesday February 10, @03:43PM (3 children)
>>It's not funny, but it's part of the answer.
>What is the rest?
Business as usual - but that's going to groan under the load of excessive detail that AI seems to put in everything it creates, at least by default. You can always ask it for a shorter summary, but at this stage I don't have confidence that it will not leave out important details while summarizing... Still, these aren't really new problems at all - people also give too much detail and leave important items out of summaries...
>I can't quantify how worthwhile the exercise is for the "vibe code review" on a "vibe coded source", the two exercises are not independent.
Neither can I. I can say that the exact same model will find faults when it reviews its own output, at least for the first several iterations. Eventually it settles down and "gets happy" with its own work, at least for things I have taken that far. I haven't set two independent models (like Opus 4.6 vs GPT 5.3 codex) against each other reviewing and revising each others' work - would be interesting to see where that process lands after N iterations, does it approach a stable result or does it oscillate? I am fairly certain that would vary from trial to trial.
>chain saws are deterministic. Cutting wood of variable quality with an nondeterministic chainsaw is not an an experience I'd dare to acquire, given the risk to life and limb.
It's not a perfect analogy. Part of why I "got into" computers in the first place (in the 1980s) was because you could experiment more or less endlessly and the worst "damage" you would (typically) do would be a full system crash / reboot, then try again. No missing fingers or limbs, no incurable diseases... of course, when you let your software out into the wild it takes on additional risk (thus the MIT license standard disclaimer...) but that's what I and my colleagues are paid for: to take the company's software "to the next level" where it's not a teenager's toy anymore, but rather something you can rely on. Is AI ready to sign off on software as "ready to use"? No, HELL NO, and anybody who tries an excuse along the lines of "the AI said it was good enough" should lose whatever license and job they had that people trusted them to make the "ready to use" assurance.
>a deterministic linter
An interesting (to me) aspect of all this is: AI is pretty good at writing deterministic parsers, probably linters and similar stuff. If you let the non-deterministic LLM review some code, then score its responses: helpful, pedantic, irrelevant, wrong/harmful - eventually the helpful range results should be able to be defined and added as modules to deterministic linters... This morning's review by Opus contained 15 observations - two were redundant-ish, and potentially valid, another for a potential race condition was also good, one pointed out a potential UI/UX enhancement opportunity, and the other 11 were of limited to zero value; about 6 limited value things were stale comments / dead code which admittedly should be corrected in best practices, and the others are basically due to Cursor's inexperience with our larger project build system - though it intuits pretty well how it works in principle, it's missing some of the actual usage practice (like default values that are overwritten before actual use) that resulted in those zero value observations.
>There have to be early adopters, but I grew old enough to no longer enjoy being among them.
At one time, I had the title Vice President, in a small company, but nonetheless there's an expectation of leadership / management / mentoring that goes with the title. Through the years I backed away from the management roles, not because I didn't like having minions do my bidding, but because people management is inherently messy and absolutely unforgiving of mistakes, in contrast with software "experiments." To an extent, AI agents behave like minions, and they absolutely do suck up to you - too much at times. It's quite a bit of fun when they "get it right" in many ways more satisfying than building the thing for myself - being able to delagate and still have it built successfully is fun, for me.
🌻🌻🌻🌻 [google.com]
(Score: 2) by c0lo on Tuesday February 10, @10:03PM (2 children)
Thank you. The discussion slightly adjusted my position from "as of today, a waste of time" into "I may give it a try, no high expectations of practicality".
The "AI is pretty good at writing deterministic parsers" and "being able to delagate and still have it built successfully" were the points.
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by JoeMerchant on Wednesday February 11, @12:13AM (1 child)
>"I may give it a try, no high expectations of practicality"
That's where I started last April or so, and I'm not convinced I have found dependable practicality yet. I have found some things it is good at (like simple parsers) - so I guess that's practical when the needs arise. But... upper management really really wants to hear how we're working with it, so I take the opportunity to use it when I can. The 2 month hiatus from early Dec to early Feb was only 3 weeks of vacation, the other 5 I was too busy catching up from 3 weeks off to really play much with the AI tools.
🌻🌻🌻🌻 [google.com]
(Score: 2) by c0lo on Wednesday February 11, @10:50AM
If you have 20 mins to spare, how accurate it this guy describing [youtube.com] the current status of vibing?
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 0) by Anonymous Coward on Tuesday February 10, @01:23AM (2 children)
For now not enough. If AI could write good enough software, Windows 11 updates would have been getting significantly better. They're not, some might say they've got even worse. 🤣
(Score: 3, Insightful) by Bentonite on Tuesday February 10, @02:16AM (1 child)
A LLM cannot write anything - all it can do is copy existing code and combinations of such and most code that exists is wrong and full of logic errors.
As a result, 99% of LLM outputs of code that isn't something as trivial as a hello world, will be code that is wrong and full of logic errors.
There have been many, many severe bugs and breakages resulting from windows 11 updates, due to how such updates were developed by using a LLM to copy random code;
https://www.businessinsider.com/microsoft-internal-memo-using-ai-no-longer-optional-github-copilot-2025-6 [businessinsider.com]
https://wccftech.com/windows-11-latest-update-is-reportedly-causing-widespread-ssd-failures/ [wccftech.com]
https://www.theregister.com/2026/01/16/patch_tuesday_secure_launch_bug_no_shutdown/ [theregister.com]
https://www.theregister.com/2025/10/13/windows_11_media_creation/ [theregister.com]
https://www.theregister.com/2025/10/16/windows_11_update_localhost/ [theregister.com]
https://windowsreport.com/windows-11-24h2-and-25h2-updates-break-iis-websites-after-patch-tuesday/ [windowsreport.com]
https://windowsreport.com/microsoft-confirms-directory-sync-failure-on-windows-server-2025-for-large-ad-security-groups/ [windowsreport.com]
https://support.microsoft.com/en-us/topic/kb5072911-multiple-symptoms-occur-after-provisioning-a-pc-with-a-windows-11-version-24h2-update-d2d30684-4e2b-47f5-9899-a00a8e0acb09 [microsoft.com]
https://www.theregister.com/2025/10/31/microsoft_has_managed_to_break/ [theregister.com] task manager launches too many processes, sucking resources
https://www.guru3d.com/story/windows-11-25h2-update-causes-unexpected-bitlocker-recovery-prompts/ [guru3d.com]
https://www.theregister.com/2025/11/20/nvidia_windows_11_hotfix [theregister.com] "Lower performance may be observed in some games after updating to Windows 11 October 2025 KB5066835"
https://www.neowin.net/editorials/microsoft-365-more-like-microsoft-404/ [neowin.net]
https://www.neowin.net/news/microsoft-finally-admits-almost-all-major-windows-11-core-features-are-broken/ [neowin.net]
https://www.theregister.com/2025/12/04/windows_11_start_explorer/ [theregister.com]
https://techcommunity.microsoft.com/blog/windowsservernewsandbestpractices/announcing-native-nvme-in-windows-server-2025-ushering-in-a-new-era-of-storage-p/4477353 [microsoft.com] windows didn't even suppo>
https://www.windowslatest.com/2025/12/19/why-you-cant-move-windows-11-taskbar-like-windows-10/ [windowslatest.com]
https://techcommunity.microsoft.com/discussions/windows11/microsoft-finally-admits-almost-all-major-windows-11-core-features-are-broken/4475930 [microsoft.com]
https://www.theregister.com/2026/01/21/outlook_freeze_onedrive/ [theregister.com]
https://www.windowscentral.com/microsoft/windows-11/windows-11-apps-like-notepad-arent-loading-what-is-error-code-0x803f8001-and-how-d [windowscentral.com]
https://www.windowscentral.com/microsoft/windows-11/how-to-fix-boot-issues-after-installing-the-january-2026-update-for-windows-11 [windowscentral.com]
https://www.techradar.com/computing/windows/microsoft-admits-windows-11-update-is-nuking-system-drives-albeit-theres-a-limited-number-of-reports-of-these-disasters [techradar.com]
https://www.theregister.com/2026/02/02/windows_hibernation_bug/ [theregister.com]
(Score: 3, Insightful) by khallow on Tuesday February 10, @01:05PM
In other words, write.
(Score: 3, Insightful) by Anonymous Coward on Monday February 09, @01:23PM (2 children)
The premise seems irrelevant to me.
Either LLMs can make perfect software or they can’t.
If they can, then anyone can use LLMs to generate the perfect piece of software to fulfill their latest need.
If they can’t, then there will always be some motivated people to create better software to meet their needs. And out of these people, there will always be some willing to open their source.
(Score: 2) by c0lo on Monday February 09, @11:55PM
The effort economics of keeping the source code free of slop (artificially or naturally generated) will poop on that motivation quickly.
It's not the creation side of the OSS that's under attack, it's the stewardship of the project; the cost of creating sub-par code just getting way lower, the maintainers' effort in vetting contribution is increased. So, what are the possible reaction?
* keep the project open and alive and thus accept the extra vetting cost
* vet the contributors, not the code - extra effort too
* close the contributors pipeline and let the project run into "coding time starvation", eventually risking death?
Where's the new point of balance?
https://www.youtube.com/@ProfSteveKeen https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Disagree) by JoeMerchant on Tuesday February 10, @02:18AM
>Either LLMs can make perfect software or they can’t.
Perfect software is a delusion. Humans don't make perfect software, and most things you may point to as examples of "perfect" software aren't perfect for any particular real world application(s).
LLMs are even more challenged in making useful software than experienced human programmers, but... they can make some stuff. People who "don't know how to code" can actually use LLMs to "make an app" that does something, simple today - maybe a bit more complex next month, and a bit more complex than that a few months later.
>If they can’t, then there will always be some motivated people to create better software to meet their needs.
Kurzweil's singularity trope is back. The LLM software companies are using their LLM software tools to write and improve their LLM software tools, and it's not a circle jerk, they are actually making progress which has been accelerating for almost a year straight now. Will it continue to accelerate? Probably, but for how long is the question - just how good will these tools be in another year, or ten years? There's plenty of breathless predictions in the press, many attributed to insiders at the ML/LLM/Software tools companies... I think the truth will be: better than today, but not progressing exponentially for very long. I don't feel like the metrics are anywhere near precise enough to even begin to gauge whether progress over the past year has been linear or exponential. I do believe it has been consistently forward, that has been my personal hands-on observation.
🌻🌻🌻🌻 [google.com]
(Score: 5, Informative) by gnuman on Monday February 09, @01:36PM (5 children)
I didn't know Amazon were vibe coders for last 2 decades.... seriously, this "giving back" is not exactly a big problem. You either participate in upstream development because it reduces your downstream costs, or you do not. 99% of OSS doesn't receive much upstream contributions and has nothing to do with people using LLMs.
Secondly, wasn't the recent issues highlighted here all about these vibe coders trying to "give back"? So which way is it?
(Score: 4, Insightful) by JoeMerchant on Monday February 09, @02:30PM (4 children)
I read an article with a similar headline a couple of months back, their premise was: because "Vibe Coding" moves so many projects into the capabilities of a single developer from what had previously been team efforts, those single developers have less incentive to open their code (to get team development participation).
Which I think is a load of bunk. Although, I must admit, when I _first_ started believing in AI agents' ability to help me with a hobby project, I switched briefly from the delusion that my project's open source license would attract development help from the world to the delusion that if I put a closed source license on it I might somehow profit. Both fantasies have no basis in my actual experiences of software devlopment for the past 40 years (and I never bothered to switch the hobby project's license off of GPL...) Your hobby project is your own, regardless of license - nobody is likely to significantly care about it unless you put 10x+ of the work into promotion of the maximum work I have ever put into coding a hobby project. If you want team participation, you need to put effort into team building. If you want commercial sales, you have to promote that while also competing in the open market. License? Virtually irrelevant to outcome for hobby level (Vibe Coding type investment) projects, in my experience.
🌻🌻🌻🌻 [google.com]
(Score: 3, Insightful) by Undefined on Monday February 09, @03:53PM (1 child)
I would mod you to +100 if I could for that.
The level of support for these is rock-bottom-low. Even putting that kind of work into promotion won't show much result, if any; because people can't be arsed to move away from (Photoshop, Word, Excel, etc., etc.) so unless your project is outright unique, don't quit your day job.
Because for anything significant, there's an end-user learning curve to (re-)climb, and the vast majority people just aren't likely to want to do that. Not to mention the fact that the number of developers who can (or are willing to) document well is such a tiny fraction that it typically makes the learning curve just that much steeper. If you could get people to read the documentation, of course.
For projects that aren't significant, it's even simpler: not enough people give a damn either way — consequently getting traction is just about impossible.
Should you manage to come up with something actually unique and valuable, that's the only time when promotion will do you any good. Until BigCompanyInc takes the idea, throws 100 interns at it, and out-promotes you. RIP Konfabulator, etc.
The only value for the vast majority of projects is in resumé boosting. Presuming you can get through the HR gauntlet to someone who cares, and of course at this point if you show up with serious coding chops, you are very likely to get triaged out by "Nah, this one's overqualified, let's get a wet-behind-the-ears collegecritter instead."
Let's not forget the NIH effect, either. You can write something really great, offer it up, and have it rejected for the dumbest (or no) reason(s.) ROI: zero.
I use a dedicated preprocessor to elaborate abbreviations.
Hover to reveal elaborations.
(Score: 3, Informative) by JoeMerchant on Monday February 09, @04:18PM
>Until BigCompanyInc takes the idea, throws 100 interns at it, and out-promotes you.
I worked, professionally, on a cutting edge software product from 1991 through 1996. Around about 1994, competitors were entering the market. Our product was DOS based, but had a mouse driven push-button / radio button type GUI. Somewhere in there was a basic, not very refined, file selector dialog. We were focused on the "science" end of things, but more than one user requested a better file dialog. Around about that time there was the product owner (a retired M.D.) + me + 1 tester / documentation developer on the project, and we got word of a competing company starting to develop a similar product - they had a team of 30 software engineers. I had to choose: add / refine a new innovative analysis algorithm, or make the file selection dialog more user friendly. With 30 software engineers those kinds of choices just aren't necessary.
🌻🌻🌻🌻 [google.com]
(Score: 2) by krishnoid on Monday February 09, @04:31PM (1 child)
Both require marketing, which, ironically (?), AI can help with in either scenario.
(Score: 3, Insightful) by JoeMerchant on Monday February 09, @04:58PM
> AI can help with in either scenario.
I agree, however... both scenarios are competitive in nature, you are trying to get support or attention or customers from a "market" filled with similar seekers. Unfortunately, those other seekers also have access to AI...
First response I expect would be along the lines of "but AI marketing is such a turn off..." and I agree, when it is done poorly. However, AI is a tool, like a chainsaw is a tool, and if you're whittling a marketing totem pole with a pen-knife you won't be very competitive, speed or efficiency wise, vs a skilled craftsman using a hammer and set of chisels, but then there are totem pole carvers who do their work with chain saws, and the good ones are very fast when compared with hammer and chisel, ridiculously fast when compared with pen-knife, but the bad chain-saw wielding totem pole carvers just make bad totem poles.
How do you "do AI marketing better"? Basically, invest more time in refining its outputs, don't accept garbage - spot the flaws and either have the AI fix them, or if that's too frustrating just fix them yourself with traditional tools.
I started "getting serious" about Claude code in July of 2025 (according to my billing history...) Back then Sonnet 4 was the latest and greatest thing. With each release since then (possible exception for Opus 4.1), the number of frustrating end-points has diminished, not dramatically, but significantly and consistently. I classified Sonnet 4 as "a helpful tool" back then, as compared to its predecessors which were barely better than a Google search. Today, Sonnet 4 looks like a clumsy toy compared with Opus 4.6 - it was still helpful, but Opus 4.6 is much more helpful, less frustrating / disappointing, far from perfect, but the models are improving at a rate I would call similar to many interns I have worked with - at least for software development tasks. I haven't used it for marketing, but I know a business owner who does and her flyers have gone from average/dull to pretty cool, interesting and venue specific/appropriate designs, in less than a year - and she spends less time making the new cool ones than she used to spend on the dull ones.
🌻🌻🌻🌻 [google.com]
(Score: 5, Insightful) by VLM on Monday February 09, @02:24PM
I skimmed the article.
1) It misses the subject matter expert situation. The value of FOSS isn't the ability to mechanically translate someone elses predefined spec into, say, C code, but being a SME enough to write your own spec, the coding is usually pretty trivial, and being enough of a SME to debug and manage and direct the continuing project. There's a LONG, I think ESR, rant from like two decades ago about that. Any moron can write code and any moron can slap a FOSS-ish license on it but only a small cadre of subject matter experts can create and manage the project, incidentally they're usually different people in commercial non-foss and usually they're the same people in foss projects. The hard part is identifying whats needed, deciding what to do, doing it correctly, evaluating, and redirecting. Very little value in "convert x= 1 + 1 into C language code" "Oh ok x=1.0+1;" (note the subtle bug LOL)
2) It misses the noise factor, suddenly instead of people on the forums shitposting "it doesn't work" that the devs ignore, its morons submitting slop PRs and issues that they don't understand to accomplish things they don't understand that generally doesn't work because god help us, FOSS engagement is a hiring and academic checkbox especially from some lower productivity countries so they just need a name in the credits for their resume and there's an infinite supply of those people. So the devs can't get anything done due to a flow of slop. "public" projects may have to go "private" or "invite only". If the only way to survive on github is to write a script to the API to click "reject" on all pull requests, why sit on public github if you're essentially not public anyway?
3) For boring stuff, NIH reimplementations might not be all that bad? I mean, someone relying on AI doesn't care about quality, obviously, so lower quality of a NIH solution isn't much of a loss, and people who want real stuff will always seek it out. No gourmet restaurant ever went out of business due to competition from McDonalds. Even the grimiest "family diner" never went out of business due to McD. Its just too low on the skill level to compete. Sure McD tilts or damages the market for real food but doesn't wipe it out.
(Score: 3, Insightful) by ElizabethGreene on Monday February 09, @08:53PM (3 children)
-1 Disagree. A code model being trained on an OSS repository costs the OSS community nothing, and as the code models get more clever, they'll enable contributions from people who otherwise would not have been able to do it before.
I strongly prefer someone submit a vibe-coded push as opposed to a flood of AI generated security bug reports because the models don't understand where I do bounds checking.
(Score: 2) by aafcac on Tuesday February 10, @01:30PM (2 children)
Anybody that can't be bothered to learn how to program now isn't going to be capable of handing in competent code with the use of AI either. If anything, using AI just makes it harder to verify whether or not the resultant code is any good. There's been tools out for ages that allow people to program by dragging and dropping code blocks and just editing a bit of the specifics like names, but that never translated into much of an uptick in the number of people contributing, but that itself had more of a chance than this does.
The issues related to understanding and tracking issues with new code are a much bigger problem than simply generating new code and people seem to just want to pretend like that isn't the case. Finding a solution is usually not that hard, knowing if it's going to be an acceptable solution without major issues in terms of security, performance or maintainability is a lot harder.
(Score: 2) by ElizabethGreene on Sunday February 15, @01:53PM (1 child)
I agre with your last point very much. Getting good enough to sniff test code has always been difficult and a critical skill. I'm curious to see if doing it constantly while vibe coding will teach that skill faster
(Score: 2) by aafcac on Sunday February 15, @04:23PM
My gut is telling me that it's going to depend a lot on how it's done and what language. Some languages like Perl lend themselves more to it due to being fairly conservative with changes and having a bunch of it to train on. It's also going to depend on people having the self-discipline to not use it as a way of avoiding doing any learning. For me personally, I don't generally use it as most of the stuff I want to know about has been done by somebody online and has an answer. AI does that a bit faster, but often times strips away the explanation and may be hallucinating.
Personally, I do think think there's some value to vibe coding potentially for one off things that solve a particular user's problems, I'd just be worried about what happens when that sort of code gets distributed. The only time I've personally used it was for homeassistant to turn my hallway light on and off based on the position of the sun as well as turning it off if it's turned on when it's supposed to be off after a delay of a few minutes. I probably could have coded it myself, but asking an AI just made more sense as there wasn't really anything particularly nefarious that could be slipped in there and the work wasn't really worth it.
(Score: 2) by corey on Monday February 09, @09:29PM
> "Our main result is that under traditional OSS business models, where maintainers primarily monetize direct user engagement...higher adoption of vibe coding reduces OSS provision and lowers welfare," the study said. "In the long-run equilibrium, mediated usage erodes the revenue base that sustains OSS, raises the quality threshold for sharing, and reduces the mass of shared packages...the decline can be rapid because the same magnification mechanism that amplifies positive shocks to software demand also amplifies negative shocks to monetizable engagement. In other words, feedback loops that once accelerated growth now accelerate contraction.”
I have absolutely no idea what this is saying, and I’m a natural English speaker, and can usually understand business buzzword lingo.
(Score: 0) by Anonymous Coward on Monday February 09, @10:20PM (13 children)
So... Apple were the original vibe coders?
The trillion dollar company that relies on LibreSSL and pf and let OpenBSD struggle to pay its electric bills?
Even Microsoft makes monetary donations.
(Score: 3, Interesting) by RedGreen on Monday February 09, @11:29PM (6 children)
"let OpenBSD struggle to pay its electric bills?"
As much as I dislike the asshole Apple corporation you could at least get your rant correct, that would be FreeBSD...
Though the FreeBSD people have a different than that take in a blurb on it I had never heard/read before, saying the roots date back before the FreeBSD.
https://freebsdfoundation.org/news-and-events/latest-news/apples-open-source-roots-the-bsd-heritage-behind-macos-and-ios/ [freebsdfoundation.org]
"Cervantes definitely was prescient in describing a senile Don fighting against windmills." -- larryjoe on /.
(Score: 2) by Bentonite on Tuesday February 10, @12:40AM (5 children)
Apple copied a bunch of things from Mach, NonFreeBSD and other BSD's, including "Open"BSD.
Apple ships LibreSSL, which is from "Open"BSD.
Of course NonFreeBSD can't help but shill proprietary software and even bootlicks so hard that they refer to devices that attack the users and takes their freedom as "stylish and user-friendly".
(Score: 2) by aafcac on Tuesday February 10, @06:12PM (4 children)
Non-FreeBSD, really, that sort of attitude is why nobody takes people like you seriously. There's more than a little irony in promoting less free licences as being more free.
(Score: 2) by Bentonite on Wednesday February 11, @03:10AM (3 children)
As I am serious, I refuse to write something that is not true.
NonFreeBSD is not free software, as it contains proprietary software that takes the users freedom - it contains many unlicensed files and many proprietary software programs without source code.
See below for just a few of the proprietary programs included in the "src" tree (included in the default install);
https://cgit.freebsd.org/src/tree/sys/contrib/dev [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/nvidia/LICENCE.nvidia [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/nvidia/tegra124_xusb.bin.uu [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/nvidia/tegra210_xusb.bin.uu [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/otus/otus-license [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/otus/otus-init [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/iwm/LICENSE [freebsd.org]
https://cgit.freebsd.org/src/tree/sys/contrib/dev/iwm [freebsd.org]
One of the reasons why BSD's use weak licenses like the 3-clause BSD, is that such permits taking the users freedom by including proprietary software anywhere (even right in the kernel).
You're confused between freedom and power.
Strong free licenses like the GPLv{1,2,3} gives both the developers and the users freedom, as you and everyone else can (or ask someone else to do so on behalf of them);
- Run the program as you wish, for any purpose.
- Study how the program works, and change it so it does your computing as you wish. Access to the source code is a precondition for this.
- Redistribute copies so you can help others.
- Distribute copies of your modified versions to others. Access to the source code is a precondition for this.
Although, strong licenses don't grant the *power* to take these freedoms away from others, to ensure that the software remains free for everyone.
While weak licenses can grant these 4 freedoms (provided the software is validly licensed and has complete corresponding source code available under the same, or another free license - so not the BSD's - only parts of the BSD's), those also grant the *power* to take any or all of the 4 freedoms - so what happens every single time, is that a proprietary software developer comes along and exercises the granted power to take the users freedom.
As you can see, clearly strong licenses are more free (as those ensure freedom), while weak licenses are less free (as those ensure freedom usually doesn't result).
(Score: 2) by aafcac on Wednesday February 11, @03:38PM (2 children)
Again, that is childish and not likely to win anybody over to your side.
(Score: 2) by Bentonite on Thursday February 12, @01:48AM (1 child)
There is nothing childish about a strong argument and backing it up the facts.
What is childish is you attacking me personally, rather than my argument (but I guess you can't criticize the argument to the same standard, as all you have are falsehoods, so what else can you do?).
I am not trying to win the too far gone - just demonstrating that I was right again and allowing people to learn something.
(Score: 2) by aafcac on Thursday February 12, @03:04AM
It's not a strong argument, it's a bunch of name calling and generally idiotic views. You're certainly entitled to have your opinions, but the name calling isn't helpful, nor is completely missing the point.
(Score: 3, Interesting) by Bentonite on Tuesday February 10, @12:33AM (5 children)
"Open"BSD got what they set out to get.
They explicitly selected a weak license and "shockingly" apple just copied the software, made it more proprietary and attacked humanity with it.
That wouldn't of happened if they just licensed GPLv2-or-later (or better later upgraded to the GPLv3-or-later or even better AGPLv3-or-later), but they don't want that, as that would mean not being able to legally carry out the degenerate acts of integrating a massive amount of proprietary software, without each case being given an excepted by every copyright holder.
(Score: 2) by aafcac on Tuesday February 10, @01:34PM (4 children)
BSD code will always have more of an impact on the world than GPL code does because of that. Not everybody is an extremist that feels the need to enforce their world view on others. Some people contribute code for the betterment of humanity without expecting anything back in return.
(Score: 3, Insightful) by Bentonite on Wednesday February 11, @04:45AM (3 children)
I'm not sure you're even on this planet, as if you were and you were able to think objectively, you would realize that weak-licensed code clearly has primarily had a negative impact on the world, while strong licensed software has primarily had a positive impact.
After all, considering how Macos and iOS and windows and almost all other proprietary software are enhanced by weak licensed code (usually the developer doesn't even bother to follow the license by attributing, as the weak license indicates the developer is weak willed and will not enforce their license) and how that proprietary software takes the users freedom, spies on the users, disobeys the users, controls the users, wastes the users time and/or empties the users bank balance, that is certainly not something that betters humanity - rather it is a case of harm against humanity.
Another example of the harm of weak licensed software is the Intel ME backdoor - billions of computers disobey their users, as those run a separate backdoor processor that run a proprietary version of MINIX, that executes Intel's malware (while Intel could have just written their own OS, they didn't need to, as a weak licensor wrote the proprietary software for them and even went so far to assist them with minimizing the size - of course Intel didn't even tell him what their plans were and maybe even violated his license by not attributing, so he only learned years later - but of course he did nothing and he was seriously proud of how many users freedom were taken).
Taking a look at most of the computers that do something useful and not something entirely harmful, like most internet routers and webservers for somewhat useful sites like Wikipedia, most mail servers, most SIP communication servers etc, you'll realize that those don't run a BSD - those run GNU/Linux (much of the freedom is deducted by how Linux is proprietary software, but that's the result of the developers treating its license like a weak license, by not enforcing the license).
While there are a handful of weak licensed programs that happen to be primarily used as free software on GNU/Linux, rather than as proprietary software, that is the exception, rather than the norm.
If you want to contribute code for the betterment of humanity, without expecting anything in return, the only license that will do that is a strong license and the best choices for a strong license is the AGPLv3-or-later, or the GPLv3-or-later, or the GPLv2-or-later.
After all, those licenses don't force anyone to give you anything at all in return (everyone can decide to keep their changes private, or choose to who to provide the changes to, which can exclude the original developer) - those just don't grant the power to attack humanity by taking the users freedom.
If you instead choose a weak license, maybe that would have a benefit if it primarily ends up being used as free software, but if it's any good, there is a severe risk of a malware author, or a malware company copying it to save a buck and using it as part of their proprietary malware as part of their humanity attacking activities.
It's incredible that you consider a strong argument, that utilizes logic and reason to inform as enforcing a worldview, but you don't consider making false claims that you know are false, as enforcing a worldview.
(Score: 2) by aafcac on Wednesday February 11, @03:44PM (2 children)
More made up nonsense. I'm not sure if it's still the case, but the entire internet, you know the place where you're posting, was built originally using the BSD TCP/IP stack with the vast majority of projects using it.
Face it, whether you care to admit it or not, the most used code is not GPL, or equivalent, and never will be. Projects like SQLite are under some incredibly permissive licensing terms that allows them to be everywhere. It's an absolute fantasy world to pretend like that's not the case. If you're relying upon a software license as a means to head off bad behavior, you've already lost as there's nothing inherent to stop the sociopathic behavior as often times the cost of writing the code is a fraction of the profit in doing so anyways.
(Score: 2) by Bentonite on Thursday February 12, @03:18AM (1 child)
If it's made up, then I'm sure you could have explained how, instead of only dismissing it in 4 words.
Even if the ME wasn't intended to be a backdoor, all of the security vulnerabilities it contained (that most vendors haven't rolled out BIOS/UEFI updates for) and likely still contains means that it works as backdoor in practice; https://en.wikipedia.org/wiki/Intel_Management_Engine?useskin=monobook#Security_vulnerabilities [wikipedia.org]
Amazing, you seriously immediately proceeded to make up nonsense?
TCP/IP is a protocol specification and there were several implementations of it prior to the BSD one; https://en.wikipedia.org/wiki/TCP/IP#Adoption [wikipedia.org]
In 1989, BSD released their stack to the public domain, which means portions of it ended up in many proprietary OS's (most of which are no dead and no longer used), but it seems that it wasn't a very good implementation, as it appears every single proprietary OS proceeded to replace large portions of it with their own implementation.
Therefore, it was *not* the case that the entire internet used *the* BSD TCP/IP stack, it was rather that many (but not all) IP implementations on proprietary OS's used for routers (like Unix's), contained a limited amount of BSD code that remained after the rewriting.
But looking at this chart of internet hosts https://en.wikipedia.org/wiki/File:Internet_Hosts_Count_log.svg [wikipedia.org] it looks like the internet was becoming popular even before such stack was released (but Unix and other proprietary OS's at the time were frankly terrible and therefore clearly never worked well as a router OS).
BSD wasn't popular as a router OS at that time either, as the litigation against BSD meant that no business would touch it; https://en.wikipedia.org/wiki/UNIX_System_Laboratories,_Inc._v._Berkeley_Software_Design,_Inc.?useskin=monobook [wikipedia.org]
What allowed the internet to become really popular was GNU/Linux in ~1994+, as Linux's custom IP stack combined with GNU's OS (that you could configure the packet routes and run RIP, OSPF and BGP implementations on), allowed for setting up internet routers that could scale as small or as large as needed, that happened to be gratis too, rather than costing a ludicrous sum (although proprietary BGP implementations and the like were still a problem until GNU implemented GNU Zebra).
Only several years after the lawsuit was settled in 1994 did the BSD's implement the required software to be an internet router (BGP etc), but BSD's were never popular as a router OS.
The reckless obsession with popularity by BSD developers is bad for humanity.
If the software is being used primarily for harmful purposes as proprietary software, that is a bad thing and therefore it would have been better if such software was never written.
Wrong - SQLite is under no license terms - it's released to the public domain - which is a problem for jurisdictions where public domain is not recognized.
SQLite is not what I would call a good database - it doesn't even have date and time types - you need to rather mungle dates with the date and time functions.
It may be very popular, but that popularity is a bad thing, as it's primarily used for harmful purposes.
There are better databases available under strong licenses, which I'd consider using instead.
What limits such sociopaths behaviors is how such sociopaths (for example proprietary software developers) are usually terrible programmers.
Throwing money at terrible programmers isn't going to magically get you functionally good software.
But, if a programmer with any skill developers functionally acceptable software and then offers it to such sociopaths on a silver platter (a weak license), for the price of $0, the sociopaths always jump at the chance to integrate it into their proprietary software and therefore make it operate at a functionally convenient level (I've found weak licensed software integrated into every single functionally convenient proprietary software program I've inspected so far).
Meanwhile, if a programmer with real skill licenses the best software under the AGPLv3-or-later, or GPLv3-or-later, sociopaths usually do not integrate it into their proprietary software or SaaSS, as they don't want to risk being taken to court and made to choose between either respecting the users freedom or removing it and therefore downgrading the functionality of the software.
Too bad the GPLv2 isn't as effective, as most GPLv2-only projects rather treat it is a weak license and don't enforce it.
(Score: 2) by aafcac on Thursday February 12, @05:52AM
Considering that none of what you've posted has any sort of actual relevance, you're fortunate that I gave you 4 words to dismiss that.
As far as it being replaced goes, I never claimed that was still the case, just that it had been the case and that much of what you're interacting with would have taken far longer to occur if everybody had to do their own stack. Not that any of what you've posted justifies your childish name-calling.
(Score: 3, Interesting) by jb on Tuesday February 10, @06:19AM (2 children)
This isn't caused by LLM users being "greedy" or anything like that. The fact is that they can't contribute anything to FOSS projects, for two very clear reasons.
First up, all code generated by LLMs is complete garbage that would not be likely to pass muster at even the laxest FOSS project.
Second and more importantly, they are not allowed to by law. To contribute code to a FOSS project, you must warrant that you own the copyright in that code in the first place. If you didn't write it yourself, you don't own the copyright in it, so you do not have the legal authority to license it to anyone else (regardless of whether or not the license provides the Four Freedoms). Any FOSS project which (knowingly) accepts contributions written by LLMs is just asking to get sued (not by the LLM vendors, but by any one of the billions of authors whose works the LLM vendors pirated to "train" their toys).
(Score: 2) by Bentonite on Wednesday February 11, @03:45AM (1 child)
There are no "FOSS" projects - "FOSS" is term of the weak willed, who cannot hold their own opinion and therefore try to be neutral between the corporate bootlicking of "open source" and the freedom of free software - but it even fails to be neutral; https://www.gnu.org/philosophy/floss-and-foss.en.html [gnu.org]
You now have been enlightened and now can think for yourself; will you stand for freedom, lick corporate boot or try to be neutral between both with "FLOSS"?
"open source" projects don't typically care much about keeping the copyright in order - only that it's good enough that a corporate is practically able to take it and turn it into proprietary software (the idea is that the corporate also hires the developer(s) - but sucked in - that almost never happens).
Yes, some "open source" projects by a business have a "CLA" policy for the gratis labor, so that the business remains the complete copyright holder (paired with a strong license, that ensures only the business can legally release a proprietary version) and of course, you can only surrender your copyright, if you are the copyright holder (and you can't surrender someone else's copyright for 3rd party software).
Free software projects do not require that you are the copyright holder of the changes, as that would be a proprietary requirement - the "NASA Open Source Agreement" is a proprietary license for that reason; https://www.gnu.org/licenses/license-list.html#NASA [gnu.org]
What is required is that you are either the copyright holder of the changes (and therefore are able to license under the project license), or that 3rd party changes are under a compatible free software license, with complete copyright information, including the license, the list of copyright holders and copyright years - so it is clear that x file(s) are under y license from z copyright holder(s).
LLM's meanwhile won't include any of the copyright information of the code that is copy-pasted and even if the prompt requests the license and a license is output, that will almost always be the wrong license
Piracy requires a boat, what occurred and is occurring is a case of mass copyright infringement, seeking profit.
(Score: 2) by jb on Wednesday February 11, @06:57AM
To a certain extent, you're preaching to the choir. Note that before I added the "F" the discussion above was just OSS. But I actually prefer the term FOSS, mostly because both the philosophical motivations (free software) and the economic ones (OSS) matter.
Agreed, yes absolutely that happens a lot too and there are no legal issues with doing so ... but that's nothing like the scenario that TFS was on about.
My point exactly. Which is why any FOSS project would be crazy to accept any contribution "written" by an LLM. Doing so would be a violation of the attribution clause of even the most permissive (ISC, MIT or 2-clause BSD) licenses. Note also that under many countries' implementations of copyright law attribution counts as a "moral right" which (unlike the substantive copyright) often cannot be assigned.
I wish. It would be so much simpler if we only had to dealing with a bunch of crooks sailing around on the high seas. It's relatively simple to get away with killing such brigands. On the other hand I'm pretty sure there's no jurisdiction in which it's legal to do the same to the bunch of crooks who run the companies behind the big LLMs.