FFmpeg to Google: Fund Us or Stop Sending Bugs
https://thenewstack.io/ffmpeg-to-google-fund-us-or-stop-sending-bugs/Open linkView original on lemmy.blahaj.zone877
Comments123https://thenewstack.io/ffmpeg-to-google-fund-us-or-stop-sending-bugs/Open linkView original on lemmy.blahaj.zone
I love you ffmpeg
Its insane just how important it is and the vast majority of the world doesn't even know it exists. Truly unsung heroes (everyone who works on it).
I'm surprised nobody posted the xkcd comic. I think Randall had ImageMagick in mind (he names it in the alt text) but it applies to ffmpeg as well.
I always used to think about curl when I see that comic. Maybe less important in recent years but still a corner stone.
Curl is less important in the cloudverse?
I'll bite!
https://xkcd.com/2347/
At this point that picture should be multiple layers of precariously stacked tiny pillars making up the entire base.
Kinda like curl.
Ffmpeg has been such cool software to learn. Simple filter chains can do incredible things
Please, go on!
Well for instance you can use it to apply tranparencys or other effects using the geq filter. It applies a formula to every pixel in the input and can adjust alpha, rgb values, and gamma. You can also use conditionals in your formula and have access to the current pixels location and value, so you can apply your transforms only to specific regions if you want, or do an adjustment keyed only to a specific color.
You're talking about green screen right? :D
Have you seen this? Green screen on crack.
https://m.youtube.com/watch?v=UQuIVsNzqDk
That and more really. You could use it to do a green screen effect, but you can also use it to adjust color balance, brightness or do weirder things like swapping values between colors. It gets really crazy when you are working with full video because the time of the current frame is also available to be incorporated, so you can even do animated effects.
Another powerful filter is the convolve filter. That allows you to apply matrix transformations, which can for example be used to apply a homography matrix and adjust a videos perspective.
Could be worse, at least Google isn't opening tickets as high priority asking basic questions on how to use ffmpeg.
Unlike the Microsoft teams devs: https://trac.ffmpeg.org/ticket/10341 Really funny to go "this is a high priority ticket" as if they've paid to use ffmpeg in teams.
The last reply is great.
That is next-level snark and I'm about it
I presume that's not actually Elon Musk in the replies...
That is actually Fellon Flask!
It is not
Jesus christ lmao
I would have replied with 'high priority for whom?' or 'high priority enough to pay?'
I looked up this Zied Aouina lad and his posts on LinkedIn are...😨
The fucking gas lighting in this response
"We ran AI that may or may not have found a legitimate issue, and you're not looking into it for us fast enough. That's going to drive away new volunteers that we need"
I think it's about driving away financial sponsors, not volunteer developers. So the last sentence is "That's going to drive away people who want to give you money and make OUR product worse and our lives harder."
They do sponsor a lot of open source projects though. ffmpeg should be one of them.
If ffmpeg was not an open source project, and somebody submitted a super obscure ai surfaced bug
The bug would be fixed exactly never
I fail to see how funding them would change that
Sure, if we forget about specifics for a bit, in general terms it does sound reasonable. And they should be sponsoring ffmpeg anyway as they are using it.
However some bug reports should just not happen in the first place
If Google said, look we know we send a lot of bug reports, here's 50MM a year, go hire a team of dedicated developers to deal with our nonsense, we don't have the expertise in house to train them on this codebase. I doubt anyone would be complaining.
Nothing wrong with fixing bugs even if they are obscure if you have the time and resources.
It's common in big tech companies to have a small internal team that has the full-time job of contributing to the FOSS software they use. That is how this should have been handled. Google wants a new feature/bug squished? You've got your team that can make the change, that's literally the whole point of FOSS.
I'm aware, sometimes they also provide funding for FOSS projects. Funding seems to be the option FFmpeg would prefer based on the title (though I've not explicitly seen a quote that says this).
If it's a specialty codebase written entirely in assembly, as this seems to be, sometimes it just makes sense to pay someone else to do it rather than spending 3x as long getting someone in house to do because the expertise isn't there. Or just put a bounty on it, another common way to provide funding in FOSS.
?
I must know as much as I thought.
I thought they owned Android. Is that not true?
They do, but Android is open source, and now Google is trying to close it down.
How? Are they retroactively changing the license?
They've been moving more and more out of AOSP into their Play Services for a good while now. However I suspect OP was referring to their announcement that they'll require developer verification, and apps to be signed with a certificate they issue, for any app install on a verified device (read any device sold with the Play Store). Long story short, no more building and distributing APKs without Google knowing who you are and that your app exists.
https://android-developers.googleblog.com/2025/08/elevating-android-security.html
There is also the fact that Google used to have a public git with the beta source code of upcoming Android releases, but have recently stopped publishing that source code
https://www.androidauthority.com/google-android-development-aosp-3538503/
Not all at once, but I feel like since the beginning more and more stuff has moved to closed source components like the Google services framework. Even the launcher used to be open source and that's not maintained now in favor of closed OEM (including Pixel) ones.
slowing down AOSP releases (why Graphene is looking into other phone options). Google is also trying to enforce developer signatures on apps, which would give google the power to kill small developers on 3rd party app stores and ruin sideloading, as you would have to go through google to be verified to make apks.
these are a few example that has popped up in the past year.
I don't think so but it seems you two are mixing Android and AOSP.
Android is owned by Google. AOSP is not.
I might be wrong on this but it seems to me they're replacing in Android, the OS shipped with many smartphones, parts that have open licenses, i.e. parts from AOSP. Like they are replacing open parts of code with privative parts of code.
They don't need to, they just need to develop their own components and ship those instead of AOSP's. Bit by bit, "Android" becomes proprietary because google's components are the ones that are force-fed to the users.
They dont acturally need to change the liscense at all, despite what most people think (and would logically make sense) AOSP is acturally downstream from Android. So basically as we're seeing right now if Android doesn't want to release the source code for something they just need to not push it to AOSP. It has been over two months already and Android 16 QPR1 still hadn't been upstreamed to AOSP nor are they legally required to (they are legally required to publish kernel sources which they have failed to do).
Nope. Android phones without google are a thing. Its the default when you install the OS yourself, actually
They are like viruses that kill their host. Mindless greed with no forecast or hindsight.
It was all over the second Google changed their motto from "don't be evil".
FFmpeg has every right to ask this. Google can't expect to extract free labour from the community.
Isn't that the definition of open source seen from commercial entities
If I had an open source program that is being used by fuckers like Google, who can afford to pay but don't, and then come in and demand shit. I'd just ignore them and pretend they don't exist and continue with my life. Let them bark until they're blue in the face. But first I'd put this as the first line in the README.md “if you're a big corporation and need help, come with money. Otherwise, please don't bother me”.
Not only that they have the money, but Google is actively working to lock down their streaming platform (YouTube) against third-parties and they have basically yanked the rug for their OS platform, while adding requirements for developers to sideload.
Their entire direction is antagonistic and in opposition to the core concepts of FOSS
Yup. Fuck google.
The problem is that some small but non-zero fraction of these bugs may be exploitable security flaws with the software, and these bug reports are on the open internet. So if they just ignore them all, they risk overlooking a genuine vulnerability that a bad actor can then more easily find and use. Then the FOSS project gets the blame, because the bug report was there, they should have fixed it!
I agree that this is a problem.
"Responsible disclosure" is a thing where an organization is given time to fix their code and deploy before the vulnerability is made public. Failing to fix the issue in a reasonable time, especially a timeline that your org has publicly agreed to, will cause reputational harm and is thus an incentive to write good code that is free of vulns and to remediate ones when they are identified.
This breaks down when the "organization" in question is just a few people with some free time who made something so fundamentally awesome that the world depends on it and have never been compensated for their incredible contributions to everyone.
"Responsible disclosure" in this case needs a bit of a redesign when the org is volunteer work instead of a company making profit. There's no real reputational harm to ffmpeg, since users don't necessarily know they use it, but the broader community recognizes the risk, and the maintainers feel obligated to fix issues. Additionally, a publicly disclosed vulnerability puts tons of innocent users at risk.
I don't dislike AI-based code analysis. It can theoretically prevent zero-days when someone malicious else finds an issue first, but running AI tools against that xkcd-tiny-block and expecting that the maintainers have the ability to fit into a billion-dollar-company's timeline is unreasonable. Google et al. should keep risks or vulnerabilities private when disclosing them to FOSS maintainers instead of holding them to the same standard as a corporation by posting issues to a git repo.
A RCE or similar critical issue in ffmpeg would be a real issue with widespread impact, given how broadly it is used. That suggests that it should be broadly supported. The social contract with LGPL, GPL, and FOSS in general is that code is released 'as is, with no warranty'. Want to fix a problem, go for it! Only calling out problem just makes you a dick: Google, Amazon, Microsoft, 100's of others.
As many have already stated: If a grossly profitable business depends on a "tiny" piece of code they aren't paying for, they have two options: pay for the code (fund maintenance) or make their own. I'd also support a few headlines like "New Google Chrome vulnerability will let hackers steal you children and house!" or "watching this youtube video will set your computer on fire!"
The main issue there is that project zero, where if you ignore what Google has reported, they will just go ahead and disclose the issue.
I’m going to be the asshole here. And? If I’m not getting paid, then why should I care? It’s a hobby project that I made for fun in my free time. Unless this is my living then, I’d understand what you’re saying.
slay
Greedy tech should pay. No question about it.
They should either get GPL'd or forced to pay.
All these company execs know is exploitation, and it's hilarious to see how immature they act when they don't get their way.
You think this even shows up on the radar of company execs?
I imagine they'd be aware of it if YouTube just suddenly stopped working
Man, I loved that line about how they could shut down three Amazon projects with a single email. That small bit of leverage against these parasites is all they have.
They couldn't just make YouTube suddenly stop working.
ffmpeg is published under the LGPL license, meaning that all of the published versions are free for anyone to use in anything, as long as they don't modify the ffmpeg library.
The only leverage they have over YouTube is that they could stop allowing YouTube to use future versions. That could create headaches for YouTube if it turns out there's major security issues, since then YouTube will need to either solve them with a wrapper / sandbox around the library, or write their own library, but any existing versions in use will always be usable by YouTube.
Surely Google has the resources to fix the bugs themselves. Most FOSS projects probably appreciate code contributions more than money.
there are some teams in companies like this where management doesn’t want to account for upstreaming and some engineers are happy to open a bug report, move the ticket to blocked, and move on to something else
this would probably just lead to the corporation taking more and more of a role until they take over development of the FOSS projects they care about, which is a particular nightmare I would prefer to avoid
was upset enough when Microsoft bought Github
What are you on about? It's Google, they're not gonna just close-source android more and more with every release or something...
Oh...
😭
Embrace, Extend, Extinguish
I can't say I've ever sent a security related bug report without at least some work done trying to understand how to fix it. Surely the caliber of people working for Project Zero can do that too, otherwise hi Google I'll take one job please.
Hell, I don't submit help requests without a confident understanding of what's wrong.
Generally, by the time I have my full support request, I have either solved my problem or solved theirs.
They're profiting from FOSS, nobody is trying to prevent them from doing so, but they refuse to spend small amounts of money helping out part-time coders ... and you know why. That money is going to the mid-level managers themselves.
Do the right thing and help your company in the medium run, or pocket chump change? Yeah, easy answer.
Source: trust me bro
Middle managers don't get to pocket any of the unspent budget. That's crazy talk.
I mean, bugs are bugs. It's not like Google makes them they are there. It's up to ffmpeg to decide if they shoul care or not
But in general I think companies who rely on opensource need to contribute more.
No but there are big bugs and small bugs and it sound like Google's AI bug finder is flooding them with small bugs that don't effect the security or end product so much. But some unpaid volunteer from FFMpeg has to check them all out regardless. And Google getting pissy about it doesn't help.
The bug in this case was a vulnerability in 1995's rebel assault 2 video game cinematic, specifically the first 20 frames. So only people with that game, watching the specific cinematic, who got the special hobby build of ffmpeg, had this vulnerability.
Okay so, the same industry that is trying to kill video games is now worried that a game from 30 years ago nobody ever heard of has a bug?
Google needs to go back to taking their meds.
Yes, but still a bug. Ffmpeg could just have said "OK. We not gonna patch that "
Google also appended a 90 day disclosure policy to their reports. FFmpeg can always say , we're not going to fix that, but that would mean a security issue would be published, and letting nefarious actors act on it. Even if it would only affect 3 users, the idea that the follow up information of, "don't use FFmpeg for this use case or you'll be hacked," would be out there.
The criticism arrises from the fact Google, the multinational mega-corp, is sending these reports with the 90 day disclosure policy to a tiny unpaid team. How about the company with something like $100,000,000,000/year in net income offer a patch or two?
Sounds like a prioritization issue. They could configure the git bots to automatically flags all these as "AI-reported" and filter them out from their TODO, considering them low priority by default, unless/until someone starts commenting on the ticket and bringing it up to their attention / legitimizing it.
EDIT: ok, I just read about the 90-days policy... I feel then the problem is not the reporting, but the further actions Google plans based on an automated tool that seems to be inadequate to judge the severity of each issue.
That was an incredibly interesting read, and I learned a lot! Thank you for posting it!
It’s genuinely infuriating that so much labor is simply stolen, in so many different ways, from people with a passion for what they do, and turned into profit for some mega corp, with the vast majority funneled to a few people completely unrelated to
theany work.Anyone who doesn't work for themselves is getting their labor stolen, and that includes me. The name for this type of systemic crime is "capitalism."
Not if you are being compensated for your labour. The actual crime that describes stolen labour is "slavery"
I think you could make an argument that being compensated for your labour, but way under the value your labour produces and also under the constant threat of homelessness and starvation if you don't do it is still an unethical system.
Nothing was stolen. The authors choose to give it away, for free, with no strings. That's not theft.
No one forced them to choose that license, and no one forced anyone to contribute to that project.
I wouldn't call the L/GPL "no strings".
True. I should have been more specific. No strings in that there should be no expectation of receiving anything in return.
"Allow me to interject and explain the four liberties..." (Or, goto fsfe.org/freesoftware )
If I understand correctly the biggest issue for FFMPEG and other projects is not only the Google and Microsoft that use them without giving back, but their chosen License. They gave permission to corporations to do this. One of the potential ways to fix this situation, is to change the license. For example from LGPL to AGPL. And then they can sell the legalese package of allowing them to break their license. The biggest difficulty is that, as a project, they'd need consent from every past and future contributors. So, yeah. I get it. This is a mess.
It would be way more easier if more corporations donated to open source projects... There's too much labour that's invisible
I don’t understand what switching from a permissive license would do here.
Some GPLv2 projects monetize by selling: support, extension via custom features, or simply the permission for a commercial use. This is possible, and it's what I called "the legalese package". Imagine ffmpeg being able to charge every year any amount they want to the biggest clients, like GAFAM. Yet you're still able to use it non commercially... To be fair, there're some middle uses, that get the disadvantage of having to break the license or ask for permission. For example, if you create anything with ffmpeg, then as an indie dev you'd need to launch your product breaking the license or paying them... But even so, situation is manageable (e.g. ffmpeg could spare you and/ or give a 1 year permission to small businesses)
I appreciate the lengthy answer, I just don’t understand what this has to do with bugs.
It's unclear what you are trying to say. The question was what would switching license do. There's 2 scenarios: 1) either Google is really not doing changes in ffmpeg source internally right now ...or 2) they are in fact making changes to it internally (perhaps for encoding with their own codecs, etc.) which they are not releasing back to the public (since the code is LGPL, and not AGPL)
With situation 1, they can simply continue using ffmpeg, even if it were to switch to AGPL. They would have no need/obligation to release anything, whether they decide to fund development or not. The way I see it, only if it's situation 2, will Google be affected by a license change. However, if the use they make of ffmpeg is just to have their own encoder program for use with specific codecs, they might as well decide to stop using ffmpeg for this purpose instead and have their own program to work with their encoders. Most of the encoding work is already being done in the encoding libraries separately released (like libaom, which Google licensed under BSD-2).
But even in the rare case of Google having made changes that (after license change) they would suddenly decide to be willing to share with the community despite having not done so before... the whole problem with this bug-reporting mess is that most of the issues reported by the automated tools aren't something really that impactful/important, they are things that even Google would not really be that interested to fix.. (why would Google need to fix a codec that only affects a videogame cinematic from 1995?). These reports are just the result of automated & indiscriminated AI analysis, slop.
AGPL is more "copyleft", but not really more "permissive", in the sense that AGPL adds the extra requirement of forcing server admins to provide the sourcecode to the users of any service that internally makes use of AGPL code.
It plugs a loophole of the other GPL licenses that allows companies to not share any custom modifications as long as they don't directly share the binaries (they can offer a service using internally modified binaries, but as long as they don't distribute the binaries themselves they don't have to share the source code from those modifications running on their private servers, even if they are GPL).
However, I don't think a license change would really solve this particular bug-reporting trouble. Most likely Google has not patched these vulnerabilities internally either, or at least the biggest chunk of them (since most of them are apparently edge cases that would most likely not apply to Google's services anyway).
I mean, I understand the licenses, I just have the same reservation you addressed at the end: I don’t see how the licensing scheme would affect bug reporting.
They should just call this an incomplete AI output. If the AI is so good, it should create the fix, add tests, and ensure nothing else breaks.
Then file the bug back to them
Yeah slave, stop complaining get your ass back to work because I'm about to dump more obligatory work on your lap that you will fix for no pay, I don't care you have a family to feed!
Your complaining about not having any sponsor for the free work that we sell for millions of dollars may cause that you don't get any sponsors!
The entitlement and mental gymnastics here at display is insane
Google has made billions off of open source software they got and used for free. Sure, they gave back a few fractions of a penny for each million they made with it, they gave back with adding some softwares here and there when it strategically suited them, but the simple fact is that without open source software, Google wouldn't exist today, definitely not the way they do now.
Hell, the internet wouldn't exist as it does today, it would be a tiny fraction of what it is today without open source software. Open source software is amazing yet most people in the world don't even know that it exists, that it's a concept, and that people are doing this
Yet there are countless companies profiting majorly from the work of others without giving back a dime. There are multinationals that profit in the billions from open source software without giving back properly or at all.
We need an updated GPL amendment or something that requires companies to start giving back productively in some form or another once they start majorly profiting from the work of open source projects.
"This library comes with ABSOLUTELY NO WARRANTY"
- "But the 1995 rebel assault build tho."
This would be the simplest solution. Yes, feel free to find and report bugs - but we will fix them at out own pace and availability. The vulnerabilities will be in the open and exploitable until we get to fixing them. If you need it faster, you can contribute money, people or patches.
Has anyone read the article? I barely understand what the fuss is actually about, the text is meandering and repeats semi-relevant details (specifically the part about libxml2).
In a nutshell:
Google is spending a shitload of money to find bugs in FOSS projects, but then refuses to spend the fraction more it would cost to contribute an actual fix, rather than just a bug report.
Basically, they are willing a spend a ton on finding a bunch of work for FOSS developers to do, but not on actually getting any of it done.
Not just that the bug they reported only affects some obscure LucasArt codec which isn't even included in the build by default. Plus I'm pretty sure Google heavily uses ffmpeg for YouTube.
Plus google doesn't really care if the obscure LucasArt codec is actually fixed, they're raising the bugs publicly to sell their AI. This is marketing, not security. The more bugs it finds the better, since sales doesn't care about the quality of the bugs found.
I read the article, and the title is a pretty decent summary. AI is being used to find a never-ending supply of bugs (a number of which are trivial at best). The issue that not only are the bugs being found by unlimited resourced AI, those same processes are revealing them to the public after a time. This is placing undue burden on unpaid volunteers. So "FFmpeg to Google: Fund Us or Stop Sending Bugs".
and some are, apparently, obscure af:
Great game
Great name
To add to the other replies: This is what AI is for. Not to replace labor, but to enhance the ruling class’ ability to exploit labor.
As a convenient side effect: If you use AI to spam people with bug reports, you’re basically DDoSing them… unless they then decide to use AI to help triage the avalanche. And wouldn’t you know it, Google just happens to sell AI to help you solve this problem they made for you!
“Nice FOSS project you got there. It’d be a shame if something happened to it.”
And also also: If FOSS in general turns into a ghost town… where are you gonna turn to get that boilerplate code you need to do a common task? That’s right, AI baby! All roads lead to boiling the Great Lakes so Nvidia can pay itself back.
This reminds me of that time there was a critical vulnerability in some core open source library that basically everyone depends on, and there was no one around to fix it or something. I want to say it was 2015? I can’t remember the name of the software package.
OpenSSL heart bleed, for sure
Great example of corporations just taking from open source and not giving back a dime because fuck you, give us your work!
I'd love to see a GPL version where if you're a company, and you make more than x amount of profit with open source projects, that you gotta fund it with y amount, depending on your profit or something
ALL big tech companies have gotten ginormous thanks to open source software, and though some have given back something, and some have done some funding, it's always been such few pennies on so many dollars that it might as well have been slavery. Add to that that many times what was given back was only given back because it was a good thing, strategically, for them.
Tech companies are abusive as fuck which made them so insanely big, powerful and rich and this nonsense has to stop
Open source is awesome and ALL software should be open source as far as I'm concerned, but the abuse from tech corporations has to stop
OpenSSL?
I think that might have been it.
xz?
Someone else said OpenSSL. I think that’s what it was.
There was also log4j.
Even if the license allow to use it commercially I don't think this is allow to abuse it when the only brake restricting you from donating is capitalism. These companies worth more than 3T, and they are thinking long to donate to their fondations...
With how short a time they give, if I wanted to cause chaos and previously had to do hard work to find big flaws, now all I have to do is sit back and wait for google to hand me the keys to someone else's system now.
But what will our llms do?!
They are welcome to fix the bugs themselves and make it public. Valve have done a fair bit of that with making windows games run on Linux IIRC.
They could even use their LLMs to fix the bugs, and everyone else can reject the shitty bugs it creates.
Exactly my thoughts, give the devs access to your wonderful LLM's and a decent server to help fix the issue. Google kinda behaving like an entitled first day Stack user.
Next weeks headline
FFmpeg to Google: Please stop submitting these shitty LLM created pull requests
Yas Queen
I haven’t read it yet so maybe this opinion may be slightly off topic but I think there is nothing wrong Google Sending bug reports. It only gets fucked when they actually request features
Google spent money to find bugs but won't spend money to fix them. That simply makes the devs' lives worse. It's an asshole move.
I agree.. I mean they are not forced to fix the issues, if the issue is obscure and not many people are affected, then there's no reason why they can't just mark it as "patches welcome" and leave it there. I feel this is a problem in the policy the project might have for prioritization, not really a problem in QA / issue report.
For context:
To me, the problem shouldn't be the report, but categorizing it as "medium impact" if they think fixing it isn't "a valuable use of an assembly programmer’s time".
Also:
Would it be truely better if the issues wouldn't be reported? what's the difference between the issue not being reported and the issue not being fixed because it's not seen as a priority?
Triaging and investigation take time. Plus having a bunch of open security issues even if they’re not critical destroys public confidence in the software
Sure, but if it wasn't triaged why consider it "medium impact"? I feel when tight on resources, it's best to default to "low priority" for all issues whose effect (ie. to the end-user, or to the software depending on it) isn't clearly scoped and explained by the reporter. If the reporters (or those affected) have not done the job to make it easy to quickly see why it's important to have this fixed then it's probably not so important for them to have it fixed. Some projects even have bots that automatically close issues whenever there has not been activity for a certain time (though I'd prefer labeling it / categorizing as "low engagement" or something so it can be filtered out when swamped, instead of simply closing it).
About "public confidence", I feel that this would rather be "misplaced confidence" if it's based on a number that is "massaged" to hide issues. Also this is an open source project we are talking about, there isn't an investment fund behind it or a need for people to have absolute loyalty or blind trust. The code is objectively there, the trust should never be blind. If there wasn't a long list of reports I'd be more suspicious of a project as popular, frequently updated & ubiquitous as ffmpeg. Specially if they are (allegedly) not triaged. Anyone who decides to choose ffmpeg based on the number of issues open without actually investigating from their end how relevant that number actually is... well.. they can go look for a different software.