On one hand people tend to side with youtube-dl, because everybody claims they did nothing wrong. But it can be argued that the testsuite (and CI pipeline) is the responsibility of the person who implemented. And I agree with that point.
From youtube-dl's perspective, this is an unnecessary attack surface that can be easily removed. I mean, this was bound to happen at some point, don't kid yourself.
With a project so big, I think it is irresponsible to not care for its legality. I mean, they could have just prevented this with a supersimple fix.
0. Use only CC licensed content in testsuite. If necessary, open up your own youtube-dl-tests channel and upload a video of your dog or cat. It just takes less than an hour.
1. Implement a license check, required by all plugins. Default to proprietary if none given. If proprietary, provide a url to Terms Of Service.
2. Let the end user manually confirm that the license is XYZ once it is proprietary and not public domain or CC based. Do this with every download.
3. Introduce a CLI flag like "--agree-to-license" in order to skip that.
4. Document this under legal compliance in the readme, and state that youtube-dl does in good faith try to comply with legal requirements, but that the responsibility is the action of its users that it cannot influence.
Boom. Plausible deniability, and RIAA would have not a single basis for the case.
I mean, this would have been so easy to implement. I don't understand why this wasn't integrated.
0. As people have stated before, the whole point of those tests is that they test a specific kind of obfuscation that YouTube only uses for certain partner videos. So unless someone manages to dig up some obscure partner video that has that feature turned on and yet arguably consists entirely of freely licensable or non-copyrightable content, that won't work.
1~4. youtube-dl isn't just a user facing tool, it is a library and tool that lots of other software shells out to, e.g. so you can play YouTube videos from your favorite standalone player in better quality than using a web player. Integrating ToS links and extra user interaction into that isn't trivial.
As I said, it was a suggestion to remove that attack surface. If you come up with reasons why it is impossible to comply legally, then don't cry when the legal system f*s you up.
Legality is not an option. It is a necessity, by definition.
I think there are essentially two problems here that decrease the signal-to-noise ratio in these kinds of discussions.
One is that people often mix political and legal matters in colloquial discussions of such topics. I frequently catch myself doing it too.
It is clearly not in the interest of most people for functionality that youtube-dl offers to be illegal. In fact, most of the time it isn't really illegal, in a de facto sense, since it is never acted upon and prosecuted.
However, sometimes a shady organization such as the RIAA acts upon this legal grey area and tries to push the needle in its own direction. When people then say "There is actually nothing wrong with this, this is definitely not illegal.", they are actually making a political statement without realizing. They are really saying: this shouldn't be illegal because it makes no sense that it is so. Since people actually hold all the power, this act makes this the reality in a way, but in order for this to completely materialize, the will of the people must first be focused so that it is reified into law.
The other problem is that the world is not the US and there are many jurisdictions where circumventing copyright protections and/or downloading a media file for personal use are not problematic at all. This fact also often seeps into the discussion, but sometimes only implicitly, so people often talk past each other.
Overall, in this case you cannot assume to be right by relying on fair use, because fair use as a legal concept implies that there has to be a court decision in order to exist.
And all of the required measurements in order to apply fair use are almost impossible to prove, therefore it is an unlikely case to win in court.
I agree with the political sentiment in this discussion, and I have a biased opinion, too - but when talking about pushing the needle in a specific direction the maintainers of youtubedl did a bad job at doing so.
I mean, they have one of the most powerful industries siding against them, with nearly unlimited funding, so it is obvious that this was bound to happen and they should have been more careful.
Especially when they, as a project, cannot afford lawyers to fight potential battles in court.
That's why I described it as an unnecessary attack surface, because that's what it is.
> they have one of the most powerful industries siding against them, with nearly unlimited funding
The music industry through RIAA are aggressive and may be punching above their weight class in litigation and lobbying, but remember that they are not really that large in the grand scheme of things. Certainly not in the class of juggernauts like oil, medicine etc.
Although this is fact is probably correct, and is mildly interesting, it has no relevance here because $11.1 billion is certainly a hell of a lot more than youtube-dl has to fight them in court.
That's just a reason to not pay them anything ever even you find it desirable, they are obviously incapable to handle the money, powerful and incompetent is just too malicious.
How does an automated test suite that downloads a mere few seconds of data simply for the purposes of software testing constitute infringement and not fair use? We've allowed this to go way too far. There is no way the RIAA can prove any sort of damages from this. This is an abuse of the copyright system for the sole purpose of attacking a tool they are unhappy with. Given the RIAA's history, are we in any way surprised by this?
It is extremely important to recognize in all this that as a country formed during the enlightenment, the founders of the US understood the importance information sharing holds for a free society. Copyright is intended to protect against wonton plagiarism and illegal redistribution. Fair use provisions exist to prevent abuse of this system and allow for legal redistribution and derivative works. These concepts. Patents and copyrights are at their core, systems designed to encourage distribution, not restrict it. Groups like the RIAA, Disney, the MPAA, and others are chipping away at this.
I was quite surprised by your claim, as I'm quite sure it is a right in EU and other countries. It seems you have been correct, though not any more. See quote[1]:
The U.S. Supreme Court has traditionally characterized fair use as an affirmative defense, but in .. (2015) .. U.S. Court of Appeals .. concluded that fair use was not merely a defense .. but was an expressly authorized right, and an exception to the exclusive rights granted to the author of a creative work by copyright law.
Fair use is still a defense to copyright violations, not a right. The 9th Circuit did not hold that fair use was a right in that case (the dancing baby case), only that a copyright holder must consider whether the alleged violation was potentially fair use before filing a lawsuit.
Importantly, the copyright holder is only required to consider whether the violation is fair use. They can still decide that it is not, and proceed with their lawsuit, with the only additional effort required being to state in their claim that they considered fair use and decided that the violation was not fair use.
(Note: while the 9th Circuit precedent is limited to the West Coast of the US, because most tech companies are HQ'd in CA or WA, so as a practical matter this case governs the internet.)
I'll quote from the first paragraph of court's opinion:
We hold that the statute requires copyright holders to consider fair use before sending a takedown notification, and that failure to do so raises a triable issue as to whether the copyright holder formed a subjective good faith belief that the use was not authorized by law. [2]
Note that the case was about abuse of DMCA takedowns by copyright holders if not considering fair use. Judge considers fair use as authorized by law, which I would interpret as right, not defense.
Another quote:
We agree with the district court and hold that the statute unambiguously contemplates fair use as a use authorized by the law. [2]
Fair use is not just excused by the law, it is wholly authorized by the law [2]
From previously quoted wikipedia article, the paragraph continues:
"Fair use is therefore distinct from affirmative defenses where a use infringes a copyright, but there is no liability due to a valid excuse, e.g., misuse of a copyright." [1]
As for plaintiff considerations, IMO, of course plaintiff is required to argue it's standing, otherwise the case is thrown out, or in this case DMCA is abused. Normally the first bar is to make a coherent argument, then the court gets to decide whether your opinion is right, but that does not lower the defendant's rights to a "defense", even if they are a defense.
You mostly just restated what I said. The decision merely requires the copyright holder to consider fair use. It doesn't require them to accept that the alleged violation is fair use or even to do a detailed analysis:
In order to comply with the strictures of § 512(c)(3)(A)(v), a copyright holder's consideration of fair use need not be searching or intensive. We follow Rossi's guidance that formation of a subjective good faith belief does not require investigation of the allegedly infringing content
As for the quote, Fair use is not just excused by the law, it is wholly authorized by the law
The court is referring to 17 USC 107. Fair use exists in the context of whether something is a violation of copyright, in the same sense that self-defense is authorized by the law but is still a defense to criminal charges. Notably, I didn't call it an affirmative defense because it is not one.
Then, you can't test that and former youtube-dl maintainer said in an interview that they download just few seconds of it and he believe it satisfy the fair-use but well, he think it should be removed to avoid issues if necessary.
>If you come up with reasons why it is impossible to comply legally
It isn't illegal. Copyright protects the video/audio content, not the hyperlink. You would have to establish ytdl induces infringement. Grokster induced infringement. An automated test does not induce infringement.
I am not a lawyer, but my impression is that the distinction is important. If you can demonstrate that those links were chosen because there were no viable links that were free of copyright, rather than to encourage users to pirate things, it would strengthen a fair use argument.
Many people are pointing to those links as evidence that youtube-dl was intended to be a tool for piracy (under the assumption that downloading videos from YouTube is automatically a DMCA violation, which I'm not fully convinced of).
> Legality is not an option. It is a necessity, by definition.
Publishing the software anonymously on hidden servers (say, Tor onion Services) is not an option. It is a necessity, by definition, because the legal system will f*k you up.
>As people have stated before, the whole point of those tests is that they test a specific kind of obfuscation that YouTube only uses for certain partner videos.
what is the purpose of this obfuscation? I mean if it is to hide the licensed content then legally speaking it doesn't sound like a great argument.
I didn't dive in the code that much but I think it is a cost effective DRM, a protection you can't remove accidentally by renaming a file on your computer (ie you need to break it specifically even if removing said obfuscation is easy) and thus is protected by DMCA.
In my opinion, it is the only argument in RIAA letter that is relevant.
So to have a "nice" youtube-dl, just remove those tests and more importantly the DRM removal (not sure it will benefit RIAA in long term though...)
It's not DRM. It can't be DRM, obfuscation is never a valid measure to restrict access. The purpose of obfuscation is the opposite, it makes sure that you can still for example run the obfuscated code, but it's just hard to understand what's going on.
The reality is that there is no "circumvention of protection measures", like RIAA is claiming. Youtube provides a javascript to decrypt their cipher, youtube-dl runs it just as intended and that's it, that's just what a normal browser does.
It can't be DRM, obfuscation is never a valid measure to restrict access.
From a technical (code) perspective, obfuscation may be an inadequate protection. But legally, the DMCA only requires some technical measure, and obfuscation satisfies that requirement. It is irrelevant that youtube-dl waits until after Youtube decrypts the content, since the law as written is agnostic to the specific technical details of how copyright protection schemes are implemented.
It looks like a measure against hotlinking, but not against execution of the script as intended, if that's what youtube-dl does. Is it what they call winedive or something else?
The content itself is not encrypted in any way as far as I know. They encrypt some kind of URL, to the actual content I think.
If this is truly the case, then every web browser that can access YouTube and supports JavaScript is guilty as well.
It's like distributing an encrypted video along with a python script that decrypts it and then saying that anyone who executes that script on linux is circumventing the copy protection. It's ridiculous.
No, that's not how the DMCA or copyright works at all.
Youtube-dl will not win this argument based on technical details on how content is provided by a licensee to an end user. That's irrelevant to how copyright law works. The entire history of copyright law has been about preventing the unauthorized copying of protected works through technical means, so copyright laws are deliberately agnostic to how the technical details are implemented. In the case of the DMCA, all that is required is that some technical effort went into restricting the unauthorized copying of licensed content.
A lot of HN commenters have pointed out that the technical efforts were quite simple: obfuscating the name of the cache file in the browser's cache folder. But for legal purposes that is sufficient. And if it's not, then in the very near future Youtube will no longer store RIAA music videos (or possibly any content) in an accessible cache folder, and congratulations: youtube-dl (and its competitors) won't work at all for it's namesake site.
I mean, sure, but the argument is that it circumvents the protection measures. If you willingly give someone a dedicated tool that decrypts your content, I just don't understand in what world using it is considered to be a circumvention of protection measures.
I have no idea how American courts usually work, but what a given program actually does seems pretty relevant to me, if you allege something like that. But I guess that's just my opinion and the American law really is that fucked. Thanks for your answer.
If you willingly give someone a dedicated tool that decrypts your content, I just don't understand in what world using it is considered to be a circumvention of protection measures.
No, they don't willingly give someone a dedicated tool that decrypts the content. Youtube simply does that for expediency.
The end result if youtube-dl wins on these technical access grounds is that Youtube will be forcedby the RIAA and other content licensors to stop caching content in accessible form. Youtube-dl will win the battle and lose the war.
The point is that they provide the tool for clients to decrypt the URL. And youtube-dl is a YouTube client, just like Chrome or Firefox is. If you have a key to a locked house, unlocking the door doesn't mean that you "circumvented the door" and went in through a window. I really don't know how else to put it.
I might be missing something, but I'm not sure what's the caching thing is about. YouTube tells you under what URL you can access the specific video or audio file, you download it and that's it. Whether the downloaded data is saved to a file or temporarily stored in RAM or whatever is up to the client. And the way everything is structured right now, I really can't see any other way they could do it without creating some kind of proprietary YouTube client.
Surely Fair Use exceptions and DDA provide legal reason for downloading any media in USA. A particular download might lie outside those exceptions, but it's not a thing you can deduce from code, you need to know who is running out and what the context is.
This comment is my personal opinion, not legal advice, and in no way relates to my employment.
The keyword here is "isn't just". It is, but there's more. For example, I use the commandline tool youtube-dl to download videos, but I sometimes also use mpv to play YouTube videos transparently (it invokes youtube-dl internally). The former use is user-facing, while the latter one is not.
These videos were never fully downloaded anyways; they are automated test cases where the test just downloads the first 10KB, which amounts to a couple of seconds at most. This is certainly fair use.
The issue isn't how much of the music videos were downloaded during testing. That may or may not fair use.
The issue is that they were testing the ability to download music videos in the first place, meaning that the tool was intended at least in part to download music videos subject to copyright.
no it's not. intent is also taken into account of fair use and the intent of those test cases is to make sure their program gets around a specific counter measures youtube as put into place so people cannot directly download the media and are forced to stream it.
look... i love the project and everything, but philip is just plain in the wrong and doesn't want to admit it.
The statement is relevant regards one of RIAA's three claims; that youtube-dl infringes RIAA member works in executing its test suite.
That averred infringement is a specific claim. Though weak, it's the strongest specific claim RIAA makes.
Hagemeister's comment establishes that minimal copying is performed in the testing itself, contrary to RIAA's claim.
The infringement does not occur in the source code or test suite itself, it occurs inccidental to testing functionality of the software (that is, it is inherently functional, not expressive), the copies made fall within existing limitations to exclusive rights (§112, §117, §512(a) and (b)), and tests (1), (3), and (4) of §107 fair use.
RIAA's standing on anti-circ §1201 is separate from this question, poorly justified, and likely lacks standing.
dude... you just dug your own grave with your statement:
"Hagemeister's comment establishes that minimal copying is performed..."
bottom line... they are still downloading and copying the file, regardless of what for or how much which is exactly what the RIAA is claiming. you can cite whatever it is you want, the fact is that youtube-dl is DEAD wrong in what they are doing.
Fair use copying, in any amount so deemed, "is not an infringement of copyright". The statute lists four tests. Any one of these, or other conditions, may be sufficient.
The third test is "the amount and substantiality of the portion used in relation to the copyrighted work as a whole".
Again, Hagemeister establishes that a bare minimum copying is performed, sufficient to verify code function. A fact which would overwhelmingly tend to a fair use finding.
you can cite whatever it is you want, the fact of the matter is, that those test cases are going to seal their doom. any lawyer worth their salt can use those as evidence of them writing specific code to target and bypass security protections so they can download copyrighted works.
i can guarantee within the upcoming weeks we will see a lawsuit filed against everyone involved in youtube-dl... and guess what, they are going to lose that case.
Youtube-dl is executing code provided by Google/YouTube, for World Wide Web user agents, as a World Wide Web user agent, and meant to be accessed and run by user agents in order to access YouTube content. That is, youtube-dl's operation is entirely within YouTube's technical design and intent.
and your article means absolutely nothing. the bottom is, we need to wait and see how is thing will play in court which i guarantee is where it will be heading soon.
>test cases are going to seal their doom. any lawyer worth their salt can use those as evidence of them writing specific code to target and bypass security protections so they can download copyrighted works.
Doesn't this apply to all browsers too? A browser has to actually download the copyrighted work to play it back, regardless whether it's "streaming" it or not. Chunks of the file get downloaded and those protections will have to be bypassed to be able to play it back, no?
We know that different media players rely on youtube-dl to be able to play back videos from YouTube. Are they not allowed to be able to play back YouTube content then?
> they are still downloading and copying the file, regardless of what for or how much which is exactly what the RIAA is claiming. you can cite whatever it is you want, the fact is that youtube-dl is DEAD wrong in what they are doing.
This doesn't make any sense to me as an attack on a fair use defense.
Fair Use is, by definition, an affirmative defense against copyright infringement. To invoke it, you assume (even if arguendo) that the copying did take place. Fair Use is a legal justification of the copying, not an argument that copying did take place.
So, pointing out that copying took place and spiking the football as if that's the end of the conversation is nonsensical if you're arguing against a Fair Use defense.
i totally agree with you, however the specific tests that the DCMA complaint is referring to is testing that the program circumvents protection to make sure the program can download specific copyrighted videos.
I know this is an unpopular opinion and I probably get downvoted for this. But I stand my ground that I think the assumption of fair use for this copyright circumventing algorithm is wrong.
------
First off: Fair Use does not exist. It is an assumption, not a defined situation as it requires legal evaluation in court to apply. And on top of that: It is a regional difference and is therefore completely unpredictable whether it applies or not.
Assuming fair use on a platform with a proprietary license that even forbids downloading, while downloading proprietary content specifically to circumvent copyright protection, while fair use does only exist as a "possibibility" inside, and only inside, the US and only after a legal case was decided in at least state level court ... really?
You get sued for downloading 512kB of porn illegaly, just as you can get sued for downloading 10kB of copyrighted youtube videos.
As I said, this was just a suggestion to remove that attack surface. Another attack surface is the trademark violation, but that's a different case.
Notwithstanding the provisions of sections 106 and 106A, the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include— ...
-- 17 U.S. Code § 107 - Limitations on exclusive rights: Fair use
How can you prove that youtube dl is used specifically for, say, classroom purposes? Can you prove it? Probably not.
Is circumventing a copyright mechanism an educational or journalistic purpose? In a testsuite? ...
Even when siding with you, you would need a court case to evaluate whether fair use applies or not.
Again, as I said, unnecessary attack surface. I don't care whether you or I am right or wrong. Nobody of us can be right unless a court case has been made.
Fair Use is a circumstancial definition in the law. It is not clearly defined and needs a court case in order to apply.
Therefore I still think that the assumption of Fair Use to apply is wrong. However, the assumption of copyright to apply on the other hand is correct.
But, referring to whether fair use exists or not, I agree with you that technically it exists.
But as the rest of the very same law criminalizes by default, you can still not refer to fair use if you do not have the required-by-law evidence at hand for it to apply.
I think that fair use is almost always nearly impossible to prove, that's why I wrote my statement like this. And I agree that it was worded badly.
Copyright is certainly easier to interpret than fair use, but both rely on courts.
If I claim you violated my copyright and you disagree, it will be determined in court. If I claim you violated my copyright and you claim it's fair use, and I disagree, it will be determined in court.
Fair use is not a “right” (as in a well defined thing you can do freely) but rather a “defense” if someone claims a copyright violation. It’s defined in terms that need to be evaluated by a judge case by case (e.g., if there is an economical damage).
Actually, in Lenz v. Universal, Fair Use is more than an affirmative defence:
we hold — for the purposes of the DMCA — fair use is uniquely situated in copyright law so as to be treated differently than traditional affirmative defenses. We conclude that because 17 U.S.C. § 107 created a type of non-infringing use, fair use is "authorized by the law" and a copyright holder must consider the existence of fair use before sending a takedown notification under § 512(c).
As another comment pointed out, youtube-fl is not _circumventing_ the mechanism, it is merely using it as intended by its creators. It does exactly what a web browser must do to play the said videos. Just like a browser is allowed to do the same thing to get the right video, put it right there in plain data on your computer and play it - no proprietary DRM ever steps in the process! - , youtube-dl follows the same steps. So youtube-dl is no different from a browser that also has a "download to your computer" button - every other behavior is the same.
"Educational" isn't a blanket license to copy. Otherwise there would be no textbook publishers, or companies that make films for classroom use.
(I have a friend who makes a living as a choral composer. Churches, he finds, are the biggest thieves! They believe they don't have to pay for music because they're a Church.)
Universal's sole textual argument is that fair use is not "authorized by the law" because it is an affirmative defense that excuses otherwise infringing conduct. Universal's interpretation is incorrect as it conflates two different concepts: an affirmative defense that is labeled as such due to the procedural posture of the case, and an affirmative defense that excuses impermissible conduct. Supreme Court precedent squarely supports the conclusion that 11331133 fair use does not fall into the latter camp: "[A]nyone who ... makes a fair use of the work is not an infringer of the copyright with respect to such use." Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 433, 104 S.Ct. 774, 78 L.Ed.2d 574 (1984).*
-- Lenz v. Universal Music Corp., 801 F. 3d 1126 - Court of Appeals, 9th Circuit 2015
It's wonderful to have a hivemind that rejects factual misstatements. Fair use does exist, in law. It's the only reason we use that particular, awkward phrase.
I agree that they should have stayed away from copyrighted material as if it were radioactive, but I'm not forced to support every bad argument that agrees with my conclusion.
As a user, I would really like youtube-dl to not break on random videos. I have a number of machines that can't handle watching youtube directly in firefox but using youtube-dl with mpv works fine. If RIAA can't handle people watching their videos then they should go build their own hosting site and force people to download some .exe DRM to watch them.
The RIAA should provide me with a dedicated device and pay me for any and all traffic use that is unrequested/ad-related, instead of trying to hack the PC and network connection that I am paying for to force me to watch ads.
Maybe the venn diagram of "people who create wild things and throw them out there" and "people who are legally cautious and circumspect" might show only minor overlap. It's sort of the opposite of creativity.
It wouldn't surprise me if a fork popped up and did most of these things right. It would also be the right time to drop the 'youtube-dl' name for something better.
For me the first thing that comes to mind for "multimedia Swiss army knife" is ffmpeg. I think that right keywords here are "web", "media" and "download" (although it can do things other than downloading, like extracting stream URLs or just querying available formats).
I hope that some large company chooses to challenge this and force it to a court ruling for precedent. (hopefully against RIAA, obviously, is my position)
But it brings up a side point in my mind -- how interesting is it that sometimes, a company, or university will have someone in their administration who cares enough to say, we're going to take this on, and not roll over? Or be willing to say, "even though this is a small annoying cost to us individually, this nuisance is taking $B across everyone and should be stopped".
That's a very rare thing to see. I remember only a couple of times this has gotten to the news in a major way:
-- When Newegg took those patent trolls to court
-- When Caltech, or was it MIT/Harvard took the MPAA to court to say that they are not responsible for their students' hosting of content
I.e. not some non-profit or foundation dedicated to the topic, but an actual business that goes out of its way (and at cost to itself) to fight something. I'm sure others can give better examples. (And perhaps I'm not aware, but some industry association also helps back the lawsuit financially?)
But still, it takes some kind of courage of legal counsel, probably with backing of the CEO or president of the company, to take stands like this and actually make it a big deal and push back.
Funny how personality and individuals at corporations can still show through what you would think is otherwise bland legal daily life. I almost take that as more of a brand equity enhancement / reason that a company gains in reputation than sometimes its products. Or maybe it's those companies that have strong culture and brand that choose (or have the strength) to do this.
> I hope that some large company chooses to challenge this and force it to a court ruling for precedent. (hopefully against RIAA, obviously, is my position)
Probably, EFF would do that.[0]
FTR, EFF is the Electronic Frontier Foundation:
> We're the Electronic Frontier Foundation. We defend your civil liberties in a digital world.
And just as a reminder, for most high paid tech workers employing specialists to do the lobbying and suing for you is the most efficient [1] way to do your part in changing the political landscape. Speaking from my own experience, I find it extremely satisfying to know that I'm actually doing something, instead of just being outraged about the state of the world.
[1] Under the "time equals money" assumption. If political activism brings you joy or counts as leisure time, the trade-off might change.
Two more examples that jumped to mind immediately:
University of California vs Elsevier (academic journal publisher), over wanting free open access for the public to their published research.
Epic Games vs Apple, over wanting developers & users to have more options and freedoms on a closed platform.
I think it’s really special when an organization is able to say “yes this impacts us specifically in a small way, but it impacts the greater community in a much larger way — we have the resources and will take a stand.”
Whether you agree with the organizations or not, I just think the act of doing something like that is special.
I don't think I'd group Epic Games vs Apple in that category. Epic Games has got to be one of companies feeling the most individual impact from how Apple runs their store. Their case is very much a 'them' thing and not some sort of public service.
What's good for them isn't necessarily bad for us as a result of that.
Being able to run whatever you want on hardware you own should he a legal requirement let alone a company of Epic's size having to fight it out legally.
I was careful not to comment on the actual good-ness of the potential results for a reason.
All I said is that Epic Games did not file the suit purely out of the goodness of their hearts, the desire to make the world a better place, and despite their own wallets. They are doing this for themselves. It doesn't mean the results can't be good, but it does mean it shouldn't be on a list of lawsuits filed for purely altruistic reasons with no selfish motives.
It’s a little tricky because there has to be some level of selfish motives for a business with many stakeholders to take a gamble like that.
Like, the UC system was paying extraordinary amounts of money for their access to that publishing platform, and the platform wanted even more for the privilege of making the articles free to the public. There was a selfish incentive there too (10s of millions of dollars). Even though they had plenty of money to just pay the fee, they could argue that they were financially benefitting themselves while also helping the entire academic community. I don’t think businesses/organizations would ever do something totally altruistically without some inward benefit.
I think the math for Epic Games leans more selfishly in their lawsuit than many others, but I don’t think the motivations were entirely selfish either. It’s possible that I’m wrong though.
Most big tech companies today try to build their own "walled garden" and are against interoperability. The "copyright" excuse is often used to shut down screen-scrapers or alternative clients (which is essentially what youtube-dl is), despite having even less legal standing (you could argue that YouTube has lots of copyrighted content that can and maybe did get copied, but in the case of alternative clients for social media, the majority of people - who are the original copyright holders for their content - would not care about which client is used to access the content they post). Defending youtube-dl would - if successful, set a legal precedent for the entire industry, and if unsuccessful would still prevent this company from ever using that same reason to shut down alternative clients targeting their own service.
Furthermore, if you are in any way funded by advertising or "engagement" (which is again, the majority of big user-facing tech today), I don't think messing with the RIAA or MPAA is a good strategy. Your advertisers depend on them more than on you, and if the RIAA informally "suggests" them to go to a different ad publisher they totally would.
No major IT company is going to get anywhere near this. Successfully defending this wouldn't really benefit the company (it's not like winning this upcoming lawsuit would absolve you of copyright, so that problem will remain - plus if you need to license content the now angry RIAA will be less likely to give you a good deal), might actually harm the company's own attempt at a walled garden, and would piss off a powerful industry that can influence your only funding source.
> but an actual business that goes out of its way (and at cost to itself) to fight something.
I think it won't be an overstatement to say that RIAA and MPAA are the scourge of the online services industry. If you host any kind of UGC and your product gets any semblance of popularity, it's only a matter of time before they try to come after you. It then comes as no surprise that many people in the industry hate these organizations with burning passion.
Wouldn't it be better to crowdfund a suit for a knowingly false claim? Or lobby the powers that be to improve things to remove "knowingly" from the text?
What do you mean "run and hide"? Developing tools to protect yourself from this nonsense is running and hiding? Is it only winning if you win the hard way? Was the development of BitTorrent running and hiding?
I think the parent's point is that instead of depending on tools that are hard to shut down, we should be working to change the legal climate that makes them necessary.
But git itself is not decentralized, but rather replicated. There's always master repo from which others fork and merge, that master repo will be always the kill spot.
I hit F12 and see my browser downloading things in the developer's network panel. My browser's disk cache and temporary directories save the files by placing them physically on my hard disk.
Are only gui based webbrowsers with large market share the only legal way to access multimedia content?
Lynx is a browser for the terminal, am I not allowed to open up multimedia links present either with an external program that displays it in a console framebuffer or converts it to ascii art?
What about mpv which uses youtube-dl in the background to play a video?
What about a container/plugin that uses mpv to play a video in a webbrowser?
> Are only gui based webbrowsers with large market share the only legal way to access multimedia content?
I believe they are the only supported way to access content on YouTube. As for whether the unofficial ways to access YouTube are legal, that depends on whether they circumvent a “technological measure that effectively controls access” [1] to copyrighted works. That’s not so clear, but YouTube does obfuscate its code to make the task of tools like youtube-dl harder.
> That’s not so clear, but YouTube does obfuscate its code to make the task of tools like youtube-dl harder.
Not for tools like youtube-dl, but for tools that let user manually extract links, such as viewing the source code of the document. Because YouTube also provides an algorithm to deobfuscate it, which humans can't run in their heads, but tools, browsers can.
These are all technical workarounds to a non-technical problem.
Ultimately, you can capture media directly from the media output. If you build software or hardware that make the process easy for a significant number of people, the lawyers will come after you.
In several European countries until 2003 or so, this was indeed the case.
For example: Making a local temporary copy in a cache or on a hard disc counted (obviously) as making a copy under the U.K.'s 1988 Copyright, Designs, and Patents Act, which had no exception for computer caches and temporary copies, and which pre-dated the invention of the WWW (albeit not of UUCP). Doing so thus infringed upon copyright.
The European Union recognized this defect in the law and came up with Directive 2001/29/EC, which explicitly carved out in article 5 an exception to copyright for "temporary acts of reproduction".
The U.K. implemented this Directive (Directives having to be implemented by legislation in member states of the E.U.) via a statutory instrument, the 2003 Copyright and Related Rights Regulations, which amended the 1988 Act to carve out this exception in U.K. law.
Browsing the WWW with caching proxies and caching WWW browsers was technically a violation of copyright in U.K. law for about 11 years. And the U.K. wasn't alone in this. Several other E.U. member states also had to tweak their copyright legislation because it didn't carve out this exception for cached WWW browsing.
But the RIAA's objection to youtube-dl is not the making of temporary copies for legitimate browsing and viewing purposes. It is the circumvention of YouTube's mechanisms in order to make permanent copies that aren't being used as an integral part of WWW browsing. The RIAA is not challenging the making of temporary copies. It is challenging the circumvention of access control mechanisms.
It is asserting that youtube-dl circumvents the "YouTube rolling cipher" mechanism, whatever that is (the referenced explanation being in German and apparently gibberish), and so is a tool for circumventing technological access control measures. So whether you can legally use WWW browsers is not the point at issue. It is whether youtube-dl is a tool that enables access to YouTube contents by circumventing a "cipher".
and many others. Interestingly, furthermore, the RIAA appears to assert that the "rolling cipher" applies to all YouTube content, however copyright licensed.
> It is whether youtube-dl is a tool that enables access to YouTube contents by circumventing a "cipher".
YouTube provides a javascript to decrypt it and youtube-dl just runs it. Unless they are arguing that using a JS interpreter that's not approved by them is circumventing the cipher then no, it's not. And if they are, it's completely insane.
> So whether you can legally use WWW browsers is not the point at issue. It is whether youtube-dl is a tool that enables access to YouTube contents by circumventing a "cipher".
But by that logic if youtube-dl circumvents a "cipher", so does Chrome and all the WWW browsers. There is no legal difference between them, YouTube provides code to circumvent a "cipher" for such software itself.
One of my favourite youtube lawyers (also tech literate), Leonard French, did a review of the case yesterday as a livestream if anyone is interested: https://www.youtube.com/watch?v=wZITscblMBA
I was really looking forward to his analysis, however he also fell into the ReadMe trap. AFAIK (and the recent DMCA/youtube-dl repo thing seemed to support) they didn't include any VEVO videos in their readme. Instead it was used as a test case.
A substantial portion of his analysis relied on the readme misinformation. It could be taken as marketing material, and there are specific provisions for when the piracy capabilities are used as marketing.
However that's not, as far as I know or can tell, actually what happened in this case. I was very interested in hearing how tests would be viewed legally, but he seems to have been working with the wrong information and so never addressed this.
Very nice, I wondered why he had the dev console open, apparently to show that the HTTP response actually uses the "451 Unavailable For Legal Reasons" status code: https://en.wikipedia.org/wiki/HTTP_451
>A countersuit agaist RIAA may help make this cartel, or others, think twice about repeating such attempts, as well as establish precedent agaist future such attemps.
A countersuit... for what? If this is pursued in civil court, they'll be up against an army of lawyers and any judgement would be minimal (I doubt youtube-dl maintainers can claim millions in damages). You'll set a precedent, although its usefulness in future cases is questionable due to the reasons listed. The only hope is for this to be pursued in criminal court, resulting in heavy fines and/or the lawyers being disbarred. However, I doubt this will happen given how cozy the RIAA is with the establishment.
The damages are going to be very small (basically just costs in responding to the notice), but section 512 f of the dmca does provide a route to sue
Any person who knowingly materially misrepresents under this section—
(1) that material or activity is infringing, or
(2) that material or activity was removed or disabled by mistake or misidentification,
shall be liable for any damages, including costs and attorneys’ fees, incurred by the alleged infringer, by any copyright owner or copyright owner’s authorized licensee, or by a service provider, who is injured by such misrepresentation, as the result of the service provider relying upon such misrepresentation in removing or disabling access to the material or activity claimed to be infringing, or in replacing the removed material or ceasing to disable access to it.
This isn't a 512 notice, so you can't sue under that. My guess is you can't get any damages because the letter is quite careful in its claims and no judge is going to find that it's in bad faith.
The article we're talking about is arguing that the RIAA shouldn't be using a DMCA 512 notice, but it seems that the are (incorrectly) using a 512 notice, or at least GitHub has interpreted their request as such.
This github comment points out that the RIAA letter seems to be deliberately breaking form to not mention “fair use”. I don’t know if that qualifies as “bad faith” but it seems disregarding fair use means this wasn’t made in good faith.
Interesting as the linked guidelines specify a separate procedure for §1201 claims, specifically requesting "what technological measures you had in place", which would seem to preclude §1201 claims on behalf of an unaffiliated third party (YouTube, vs. RIAA).
The law does not place that limitation on filing 1201 claims.
Also, the RIAA is not an unaffiliated party. They act as the collective legal representative on behalf of their members (i.e., the recording studios).
The alternative is that youtube-dl (and others) could face multiple lawsuits, one from each music company, instead of just one, and each company could test a different legal argument. The music labels can afford this. In fact, in many situations they would prefer it.
The "copyright protection scheme" in question, if it even is one, was written by and is provided by Google/Youtube, not the RIAA.
It is not even clear to me the RIAA has standing to sue under §1203: "Any person injured by a violation of section 1201 or 1202 may bring a civil action in an appropriate United States district court for such violation." RIAA are not injured due to utilisation of a non-member's mechanism.
The claim does not pass the 17 USC 1201 (a)(2)(B) test: that the technology "has only limited commercially significant purpose or use other than to circumvent a technological measure that effectively controls access to a work protected under this title".
Yes, Github (or its owner, Microsoft Corp.) may have liability under 17 USC 1201 (a)(2), "offer to the public, provide, or otherwise traffic" the code, subject to the same test above. However there is no safe-harbour provision for such violations.
Microsoft is listed on the RIAA's members page. Neither Google LLC, its Youtube subsidiary, nor parent Alphabet Inc. are. The RIAA are threatening a member for a §1201 violation against a nonmember. That's ... weird.
You can cite TFA all you want. I disagree with his legal analysis, as do most lawyers.
His first paragraph is wrong. It's irrelevant who imposes the copyright protection scheme, only that one exists. And speaking as someone who has entered into an RIAA licensing agreement in the past, our license agreement with the RIAA did require us to put into place measures to prevent copying of licensed RIAA content, and I assume that Gootube has the same terms in their license.
The next conclusion is also wrong, since the RIAA is the authorized legal representative for its members, and prior cases have already established that they can act in this capacity.
His conclusion on the (a)(2)(B) test is not correct either. The sole purpose of youtube-dl is to download content from websites which do not make their content available for download through standard means of access. In fact, the whole point of using youtube-dl is that they constantly update the code to get around the changes that websites make so that content can't be downloaded. I would bet a lot of money that this will be the deciding factor if this goes to trial.
Microsoft is listed on the RIAA's members page. Neither Google LLC, its Youtube subsidiary, nor parent Alphabet Inc. are. The RIAA are threatening a member for a §1201 violation against a nonmember. That's ... weird.
I'm not sure why techies keep bringing this up like it means something. It's irrelevant that they are a member of the RIAA. Microsoft is receiving the takedown request because they are/were hosting the youtube-dl code through their subsidiary Github and legal procedure requires the notice to be sent to them.
MS is not the subject of a DMCA infringement lawsuit. Repeat: Microsoft is not the subject of a DMCA infringement lawsuit. By taking down the youtube-dl repository, MS's involvement in this matter ended.
Youtube-DL would be the target of the DMCA infringement lawsuit by the RIAA. If they keep working on youtube-dl from this point on, it's very likely the RIAA will file a lawsuit against the maintainers/programmers.
No, they don't. The browser accesses the content directly from the website, and currently stores it in a cache folder on the user's computer. (The caching step isn't necessary but eliminates the need to re-download portions of the content.)
Youtube-dl accesses the content from the cache directory after the browser has cached the content.
Likely zero damages because nothing RIAA did creates liability for them - no false statements that I can see, just tenuous but plausible legal claims. Certainly nothing criminal there and absolutely no basis for disbarment.
You can file a declaratory judgement claim, and a precedent there would be useful. You just wouldn't get any damages.
If circumvention falls outside of the safe harbor provisions (which seems to be the plain text of the law), that just means that the ISP liability, if any, begins immediately on hosting it, and their liability for hosting it isn't negated by taking it down in accordance with the safe harbor process. OTOH, while the mental state required for a 1201 violation is not explicit, the most reasonable reading is that it requires knowledge of the circumventing purpose, which would mean liability would not attach to an hands-off distributor of material supplied by a third party until they had notice, which, is essentially equivalent to the safe harbor for infringement. So, from a liability perspective, it's at least plausible that there isn't a big difference here between circumvention and safe harbor provision.
And, on top of the circumvention, this article tries to argue against infringement that the RIAA allegations, beyond circumvention, only involve “incidental copying", but any copying not either licensed or protected as fair use (which even the article here does not claim applies) is infringement. So, the article makes the infringement case while attempting to deny it. The infringement may only be contributory rather than direct on the part of youtube-dl’s creators, as well as anyone knowingly distributing it, but contributory infringement is still infringement, and so governed by the infringement safe harbor directly.
When you watch something from Youtube in the browser, it makes copies on your computer---in memory transiently, and on disk files.
Computers aren't radio or CDs, there's no way to actually consume media on them without copies being made at some point.
Additionally, its not obvious who is infringing. Unlike say, downloading a torrent, people have legal access to youtube ordinarily, and pay for it. If you pay for offline access to music, why can't you use youtube-dl? Any infringement will only begin when your license to the content ends.
Additionally, the software just makes HTTP(s) requests and can work with any site. I actually have never used it to download anything on youtube.
> any copying not either licensed or protected as fair use (which even the article here does not claim applies) is infringement.
Not sure what you mean by this, as the article does claim the use in the test suite is fair use.
> At best, youtube-dl’s test suite may be infringing works when run [...]. Even that argument is specious: Given output is discarded, no permanent copy is retained, and the action is for research and development, and numerous Fair Use affirmative defence claims exist under §107, notably (1) and (4), test suite execution falls outside exclusive rights.
It claims that a fair use defense exists and points to two of the four factors in fair use (mistakenly referring to them as separate claims). And, sure, that's an argument that Fair Use applies, which I missed, but is not an argument that this isn't a valid 512 notice. A valid 512 notice does not require actual infringement, it requires that the person the sending the notice have and assert a good-faith belief in infringement. While we can't explore the mind of the RIAA as to whether they have such a belief, they clearly have asserted that the tool exists for the express purpose of infringing their works such that knowingly distributing it would make the host liable for contributory infringement as well as making anti-circumvention claims.
An assertion of non-infringement, whether based on fair use or anything else, is obviously a basis for a counter-notice, and also an argument to make if the RIAA actually sues the content provider after the takedown (without or without counter-notice based restoration) to the service provider.
"Microsoft is listed on the RIAA's members page. Neither Google LLC, its Youtube subsidiary, nor parent Alphabet Inc. are. The RIAA are threatening a member for a §1201 violation against a nonmember. That's ... weird."
That a really good point. This request seems ill-considered
The (not so funny) irony here is that the powers that grant RIAA the power to sue someone for providing DRM circumvention code, are the same exact powers that push for banning software encryption for common people.
Honestly it doesn't matter that it isn't breaking the law. Because even the if the RIAA can't bribe the judge and jury to liberally interpret the laws on the books, they will freely and openly bribe congress to create new laws to replace the ones that were previously not effective.
Advocates for open source software and cryptography and right-to-repair and non-draconian copyright/patent protections need to do the things that their opponents do: they need to pony up the cash to do some real lobbying. Otherwise they get stuck in the same perpetual trap of trying to defend their status, one individual situation at a time, against the overapplication of laws that were poorly conceived to begin with. Sad, but the only language that congress talks is money.
They don't need to bribe anyone. The laws already support the copyright holders.
This is really going to shock you: The Constitution explicitly grants creators and inventors "exclusive rights." (Article I, Section 8, Clause 8 ) Fair use is a statutory construct of Congress.
Fair Use is presumed in the A1.S8.C8 justification for exclusive rights: "To promote the Progress of Science and useful Arts".
Again, Not only is Fair Use presumed, but it provides the basis under which such rights should be granted.
More discussion at Cornell's LII:
Although the courts have considered and ruled upon the fair use doctrine over and over again, no real definition of the concept has ever emerged. Indeed, since the doctrine is an equitable rule of reason, no generally applicable definition is possible, and each case raising the question must be decided on its own facts. On the other hand, the courts have evolved a set of criteria which, though in no case definitive or determinative, provide some gauge for balancing the equities. These criteria have been stated in various ways, but essentially they can all be reduced to the four standards which have been adopted in section 107...
No, fair use is not "presumed." Exclusive use by the creator is presumed, since the text of the Constitution literally says "exclusive rights."
At the time the Constitution was written, society profited when people sold their inventions or creative works. And the purpose of patent law was to reward people for inventing or creating by allowing them to recoup the costs of inventing and creating.
There was no fair use exception back then; courts were extremely harsh on copyright and patent law offenders. It wasn't until Congress codified "fair use" into the law in the 1970s. https://msu.edu/~mcgrat71/Writing/Fair_Use_Rife.pdf
"Fair use is therefore distinct from affirmative defenses where a use infringes a copyright, but there is no liability due to a valid excuse, e.g., misuse of a copyright."
-- Lenz v. Universal Music Corp. (2015), Ninth Circuit.
And I hope that Microsoft does. This would be a case where the veritable army of lawyers they have could do much good -- and the RIAA would back down almost immediately, they are good at bullying small players but Microsoft?
Yeah, well, let's hope some time next week the case gets escalated within Microsoft to the RIAA contact person who can make the phone call "My man, the fuck?", with much more eloquent words of course.
Wasn’t all that long ago that Microsoft was making their own YouTube client, with downloads on mobile before Google officially released such a feature...
I don't live in the USA, why should I be affected by the RIAA and DMCA outside the USA? If I ping github.com, it is resolved as lb-140-82-121-4-fra.github.com , so what is the rational behind the global takedown?
Though the claim is made under US law, similar anti-circumvention provisions exist in international law, which is highly standardised in large part thanks to the RIAA, MPAA (video), SIIA (software), WIPO, and other copyright monopoly cartels' special-interest deep-pockets lobbying. Offshore legal safe havens are limited and vulnerable. Defence within DMCA /anti-circ / WIPO / Berne regions is unfortunately necessary. Simply hosting the repository outside US jurisdiction is not sufficient, though a valid immediate response.
From TFA.
I supect this was included by the author in anticipation of such questions.
Based on the DNS name, I'd guess that's a load balancer either in a Frankfurt (DE) point of presence, or a Frankfurt data center. If it's a POP, then it's probably just a cache to the servers in the DC, which could still be in the USA.
End of the day, Github is a US-based business, and they are subject to US laws, and thus all of their customers/users are affected by those laws, whether or not they're in the USA.
- Where the servers are located.
- Where the comparably offices are located,
- where the business is registered,
- where the employees live or travel to/from
- employees in any country who are quite happy to ship people to the US even when that employee has not broken any law.
This doesn't stand for many legal cases, including copyright, work rights, taxation, etc. I don't see why this case should be an exception with having an effect outside the US jurisdiction.
The RIAA's whole argument is based upon the premise that ALL YouTube content is copyrighted. I do not believe that is the case. YouTube-dl is just a tool. A tool can be used for both legal and illegal purposes. If a tool user does something illegal with a tool, the user should be prosecuted and not the tool. (The same logic applies to guns.)
(i) [to] circumvent the technological protection measures used by authorized streaming services such as YouTube
(ii) reproduce and distribute music videos and sound recordings owned by our member companies without authorization for such use.
Third; We also note that the source code prominently includes as sample uses of the source code the downloading of copies of our members’ copyrighted sound recordings and music videos
#1 Falls under §1201, for a maybe-anticirc tool created by a third party for whom RIAA's standing is likely nil. The anticirc use & claim itself is weak.
#2 is ... at best vague, but seems to aver contributory infringement, which ... is not related to any Chapter 1 rights or Chapter 5 penalties that I see. Users of a tool might be liable, but not its developers or manufacturers. Otherwise, a very weak §1201 claim.
#3 Fails to identify an infringement in the work but instead in its use. The infringing party given Github's CI/CD process would actually be RIAA member Microsoft. Use of works is de minimis and passes at least three of the §107 Fair Use affirmative defence tests, as well as §117 (computer program), §512(a) and/or §512(b) (Transitory Digital Network Communications).
As to YouTube works' copyright status, under Berne (US copyright since 1976), copyright exists "in original works of authorship fixed in any tangible medium of expression", with very limited exceptions, automatically, with no action required of the author. US Government works and expired copyrights are the principle exceptions. Some authorities dispute that works can be put in the public domain otherwise (I'm not convinced). While some YouTube content is in the public domain, the overwhelming majority is subject to copyright.
It is based on the weaker premise that youtube-dl is “primarily designed or produced“ or “marketed” for the purpose of circumventing copyright protection.
PopcornTime is just a tool. A tool can be used for both legal and illegal purposes. If a tool user does something illegal with a tool, the user should be prosecuted and not the tool. (The same logic applies to guns.)
Torrent clients are legal though? And prosecution has generally followed the "if a tool isn't advertising 'you can use this to circumvent the law,' it's the user that's liable for an illegal use, not the developer" principle. And as someone who uses youtube-dl to get easy copies of NewsHour... rather annoying that it was shut down.
I agree, but unfortunately the DMCA does not. In order to avoid liability, the tool has to have significant non-infringing uses, and I believe courts will look at how the tool is actually used in the wild as a part of determining that.
RIAA doesn't need standing to report on criminal activity. §1201 provides for criminal penalties for trafficking in circumventing measures. They're telling GitHub/Microsoft "This is your last chance before we call the US Attorney and file charges against you for distributing piracy tools." And the US Attorney will almost certainly lend an ear to the RIA-fucking-A.
>Further, under §512 (f) the RIAA's deceptive claim is liable for sanctions:
This is wrong, because they didn't make a 512 claim. If you read the letter, the only part that's asserted under penalty of perjury is that they're authorized to act in behalf of their member companies, which is true.
> RIAA are not injured due to utilisation of a non-member's mechanism.
They're injured when the tool is used for copyright infringement. No issue with standing here.
>Any potential copyright infringement which might occur through use of youtube-dl is at the volition of users, not the software's authors, actions would properly be directed at such users for individual acts of infringement
This ignores the fact that circumvention itself creates liability in the right circumstances.
The claim is couched as, and appears to have been read by Github as, a §512 claim, bearing several of the conspicious elements of a §512 notice. As my footnotes document, it was widely read as same (it took me some time to realise it was not a §512 claim myself).
This could be argued as intentional, deceptive, and hence fraudulent.
>This could be argued as intentional, deceptive, and hence fraudulent.
Fraud requires a false statement, generally. The letter clearly cites 1201 and never cites 512, and is missing essential elements of 512, (e.g. it never claims direct infringement). There's no way 512(f) applies.
A false representation of a matter of fact—whether by words or by conduct, by false or misleading allegations, or by concealment of what should have been disclosed—that deceives and is intended to deceive another so that the individual will act upon it to her or his legal injury.
>> Web multimedia-content Swiss army knife youtube-dl
Come on, you've already shown your bias and I need to be on guard from that point on. I love youtube-dl and built an app that used it heavily for exactly what it says in the name: downloading media from YouTube. To pretend that's not it's primary and major purpose is incredibly disingenuous.
> To pretend that's not it's primary and major purpose is incredibly disingenuous.
I'm fairly confident some people use youtube-dl more on pornhub URLs than youtube URLs, what you say speaks more to youtube-dl's origins than its current state.
My principle use of youtube-dl is through other tools for multimedia playback, most typically mpv, mps-youtube, and other command-line media playback --- video rarely interests me, and the ability to throw randfom URLs at mpv and have it find and play the appropriate media files (sometimes YouTube, Twitter URLs w/media (https://toot.cat/@dredmorbius/104565986372976258), often podcasts, other video hosting sites, Soundcloud, etc.), is quite the win.
In mpsyt I can curate search and lists for playback (painful or impossible with the Youtube Web client), mpv can take files with lists of URLs to play. Both use youtube-dl for media access
Very often audio-only for headless/background playback.
Also for offline playback when not on WiFi or hardline LAN link.
There's also considerable processing and metadata tools available, including grabbing transcripts from YouTube videos.
Youtube-dl downloads a lot more than from youtube.
I use it a more often on vimeo content embedded in webpages which usually won't play in my browser (presumably due to some privacy settings) than I use it on youtube.
It's primary and major purpose is to download videos from video platforms. One of the biggest advantages of youtube-dl is how universally it works: If a page hosts video, is even slightly common/known, and isn't primarily a warez site (youtube-dl specifically refuses to build support for those!), youtube-dl most likely supports it.
Not having to find a clean and working version of "<somesite> downloader" and install/learn it, and instead being able to rely on one piece of software is incredibly valuable.
One of the most recent things I downloaded is a Reddit video. Reddit makes the link annoying to find, but youtube-dl handles it.
I use it to watch twitch on devices that can't use the browser, and to watch videos from free TV services here in Australia in the same way, such as iview.
I do rarely use it for youtube, but given it supports so many websites, I think 'swiss army knife' is a pretty fair description.
Just because that's all you use it for doesn't mean that's all others use it for. I've seen the yt-dl issue tracker, it's swamped in non-youtube support requests.
I used it when I was a teacher to markup my students‘ videos when they submitted them to YouTube or Vimeo.
Part of the assignments were to produce content for a publicly viewable medium, and it was the easiest way to to get a certain consistency between submissions when I wanted my students to be focused on content, not files.
I also needed to have my own copy so that I had a provable record of the students work even if they decided to later take their assignments down.
On one hand people tend to side with youtube-dl, because everybody claims they did nothing wrong. But it can be argued that the testsuite (and CI pipeline) is the responsibility of the person who implemented. And I agree with that point.
From youtube-dl's perspective, this is an unnecessary attack surface that can be easily removed. I mean, this was bound to happen at some point, don't kid yourself.
With a project so big, I think it is irresponsible to not care for its legality. I mean, they could have just prevented this with a supersimple fix.
0. Use only CC licensed content in testsuite. If necessary, open up your own youtube-dl-tests channel and upload a video of your dog or cat. It just takes less than an hour.
1. Implement a license check, required by all plugins. Default to proprietary if none given. If proprietary, provide a url to Terms Of Service.
2. Let the end user manually confirm that the license is XYZ once it is proprietary and not public domain or CC based. Do this with every download.
3. Introduce a CLI flag like "--agree-to-license" in order to skip that.
4. Document this under legal compliance in the readme, and state that youtube-dl does in good faith try to comply with legal requirements, but that the responsibility is the action of its users that it cannot influence.
Boom. Plausible deniability, and RIAA would have not a single basis for the case.
I mean, this would have been so easy to implement. I don't understand why this wasn't integrated.