Piano teacher gets copyright claim for Beethoven’s Moonlight Sonata [video] | Hacker News Reader

Piano teacher gets copyright claim for Beethoven’s Moonlight Sonata [video]

youtube.com

922 pointsbitcharmer487 comments

Summary for those that don’t want to watch the video :

- Pianist creates YouTube video demonstrating how to play Beethoven’s Moonlight Sonata

- part of this video includes her playing Moonlight Sonata (of course)

- YouTube has a new ‘feature’ that scans submitted videos prior to publication to identify potentially copyrighted material

- her video is found to include copyrighted material. She is certain this is a mistake that will be rectified by disputing the copyright claim

- she agrees the only person who’s content she is reproducing is Beethoven himself

- her copyright dispute is rejected. She is found to be violating the copyright of a piece called “Wicca Moonlight”

- she can appeal the dispute but the appeal process is limited, and she risks getting a feared “copyright strike” on her account, wherein three such strikes would mean permanent closure of the account

- she says nearly anyone can file copyright claims against published YouTube videos, including bots, and is worried about all the time she’s putting in to create content being usurped by a few bad actors

I sympathise with her. I had a copyright claim on one of my YouTube videos (I’m totally small potatoes, handful of vids with dozens of views).

I made a video of myself playing my own unique interpretation of The Safety Dance by Men Without Hats on piano. https://youtu.be/ZNfMZ6g9YWo

I got a copyright violation notice shortly after posting it, by some random Latin American music company. It wasn’t clear if it was legit or not, it could have been some holding company of song rights or some failure of the AI.

I disputed the claim, saying I used no recorded material, it was entirely my own styling on the song including ragtime and stride piano influences.

Frankly I’m not even sure how copyright works on covers, particularly if style is quite different.

The company never responded in the 30 day period so the copyright claim was removed.

My channel is so small that I probably would have just removed the video. I’m not even close to the levels that are required for monetisation (1000 subscribers and 4000 cumulative view hours).

But it was a bit surprising how quickly it was claimed that I violated someone’s copyright.

Covers of songs, even if the covers use completely new styles are protected by copyright law and therefore can risk being claimed by YouTube's system:

https://www.legalzoom.com/articles/posting-cover-songs-on-yo...

I guess you need to make it a parody to be in the clear?

I'm guessing the next Weird Al has already been disillusioned with the copyright ban hammer.

Weird Al himself was well aware of this issue, and famously would only parody songs after receiving permission from the authors.

Permission from the authors is unfortunately different from permission by the copyright holders. IIRC Weird Al did it more for goodwill than legality.

> Al does get permission from the original writers of the songs that he parodies. While the law supports his ability to parody without permission, he feels it’s important to maintain the relationships that he’s built with artists and writers over the years. Plus, Al wants to make sure that he gets his songwriter credit (as writer of new lyrics) as well as his rightful share of the royalties.

https://www.weirdal.com/archives/faq/

Didn’t even need the next Weird Al. Al’s own YouTube videos were copyright claimed by his own label for violating copyright.

Not necessarily.... The Posterchild for parody is 2 Live Crew Pretty Woman. Which includes samples of the original Oh, Pretty Woman. Youtube videos get flagged for 10 seconds of a song snippet all the time.

After the Campbell v Acuff-Rose Supreme Court decision, 2 Live Crew licensed the song from Acuff-Rose music. (which is what they tried to do in the first place).

Parody is a fair use defense, so you have to go through a lawsuit to be able to assert it.

Not an expert, but I've been learning a lil about this lately. Royalties for performances of someone else's song are paid to the writers of the words and the music. The melody and words can be copyrighted, not the chords or style. So you can use the chords of an existing song and give it new melody and words, and it is your own song. Style has nothing to do it.

I love this quote from the classic The Manual: How To Have A Number 1 The Easy Way by KLF:

...the copyright laws that have grown over the past one hundred years have all been developed by whites of European descent and these laws state that fifty per cent of the copyright of any song should be for the lyrics, the other fifty per cent for the top line (sung) melody; groove doesn’t even get a look in. If the copyright laws had been in the hands of blacks of African descent, at least eighty per cent would have gone to the creators of the groove, the remainder split between the lyrics and the melody. If perchance you are reading this and you are both black and a lawyer, make a name for yourself. Right the wrongs.

Learning about the difference between a "groove", a "melody" and a song has been a wonderful rabbit hole to go down. Thanks!

I'd love to know if (the downvoters think) that has factual errors. I'm a jazz musician who will soon be releasing music online—some my compositions, some not—and planning to pay license fees etc. Thanks—I'm a totally at a loss what could have made people downvote it!

I think the problem with that post is that race (black and white people) are unnecessarily brought into it. I'm sure that musicians of all races would want this issue addressed. Making it about race is extremely ridiculous.

Ohhhh ok thank you!

The book is from the UK, I'm in Australia, it didn't seem a ridiculous thing to say to either of us. Maybe the problem was...US sensitivities. I, like I believe the KLF also, am white and have dedicated most of my life to the music of "blacks of African descent". The issue is how much more important groove is in black music, and how copyright is in origin a white thing. Changing copyright would no doubt be a massive, heroic, lifelong task. Like the way it took Douglas Nicholls[0] 30 years work to remove a racist sentence and a bit from the Australian constitution! I just don't hear calling for a hero here as a negative thing at all, the way evidently some people do.

Did any black people have a problem with the quote? I am curious.

[0] https://news.ycombinator.com/item?id=26980084

p.s. wow, it's you! I checked out your github and website (crucialflow) a few months ago, hehe small world. proof: https://news.ycombinator.com/item?id=26153386

Groove is just as important to many non-black people, so saying that only black people would care about this is annoying me. Anyway, I don't have the ability to downvote things, I can only speak for why I didn't like the phrasing.

Small world I guess, I used to be a random nobody, I guess now people have something to recognize me by. I also release music under the name Dream Scatter by the way.

> saying that only black people would care about this is annoying me.

..But I think no-one said "only black people would care about this".

Like also I think no-one said "We need black people to change this copyright system" or "this is an issue between black and white people".

I'm not sure the quote is 100% serious, if they really thought things can or will be changed in their suggested way. But if I think about it, which I didn't until people said stuff like "Making it about race is extremely ridiculous."—it just seemed a commendable but unremarkable sentiment, one I never dreamed anyone could have a problem with—suggesting a black lawyer seems apt, for reasons something to do with what could be called self-determination. It would be more awesome and appropriate if black people righted the historic wrong, than if white people did it for them because poor black people can't do it on their own. Etc.

Ok, thanks, I think I learnt something about what your and other peoples' problem was/is, and I hope you learnt something about "my side", where what is extremely ridiculous is someone having problems with the noble desire expressed in that quote. But I guess it's hard here where we don't know each other at all, and don't know about what it seems are the unshared assumptions and beliefs that made my quoting a few lines on fixing copyright from a classic book from the 1980s into something extremely ridiculous and annoying in 2021. ..Ok thanks again.

Suggesting that the KLF are racist is pretty ridiculous!

https://en.wikipedia.org/wiki/The_KLF

That doesn't have anything to do with black or white people. We don't need black people to change this copyright system, just people in general who want to make sure that musicians get paid. Saying this is an issue between black and white people is extremely ridiculous.

same here.

i am a big fan of film music. i made a small piano reduction of a favorite piece and uploaded it to youtube. i wasn't even mad when i saw the copyright claim within hours, since i run no ads.

i was however baffled and somewhat proud that my efforts were not in vain, and that the algorithm thought my poor interpretation was close to the real thing.

i do not defend the algorithm though.

You are not allowed to create derivative works from another person's creation, regardless of whether you are making money from it. You can't go and make a Shrek III flick, for example, or write a book set in the Harry Potter universe. Likewise, you can't adapt music from a film (IANAL, there are things like fair use, but in general this holds).

https://www.copyright.gov/circs/circ14.pdf

> You are not allowed to create (free) derivative works

I think putting such constraints on creative work is a cultural disaster.

Then your beef is with the United States, not with Google.

I think misappropriating someone else's work to make a buck is a cultural disaster, but then I work for a living and expect to get paid, so what do I know? Fair Use does exist to provide an avenue for the reuse of the work of others for criticism, parody, and the like., but straight up lifting someone else's work and using it to your own ends, even if noncommercial, is wrong.

Explain why it's wrong when there's no money involved (e.g. fan-fiction and similar works).

Even if there's no money involved that can still damage the market for the work it derives from. It's one thing if the derivative work consists of original characters in the same setting, or existing characters but in a what-if scenario, but as soon as you start trying to write derivative works in the same vein as the original then you're damaging the original author's ability to market his existing work, and to continue that work himself. You are, in a very real sense, taking something that isn't yours.

That's great if you ignore the fact that almost all our ideas are borrowed. Copyrights work against the movement of ideas - the ground culture springs from. It is artificial scarcity, like DRM, unnatural in the domain of information.

> as soon as you start trying to write derivative works in the same vein as the original then you're damaging the original author's ability to market his existing work

That seems ridiculous. How would expanding literature hurt it? It would be like free PR for the original. Art (tales, legends, stories, music and potry) used to be orally transmitted, we forgot that too, with every retelling it was reinterpreted and adapted. Even Tolkien didn't invent the elves and Rowling didn't invent sorcery.

depends on the derivative. I can't make a story about hogwarts. But I can make a story about a girl with a scar who gets enrolled in a magical academy. And I probably just described 20 works with that description. But I think the legal definition of "derivative work" isn't that loose.

The more gray areas for derivative work is stuff like fan art, translations, and remixes. All are technically in the area to be copyright'd, but at the same time it also helps publishers in essentially being free, natural advertisement.

I don't get why you're confused. You violated copyright and got a legitimate claim against you.

This exact thing has happened to most YT creators at this point. It's just part of doing business on YT. Some small section (<10%?) get three "copyright" strikes (nothing to do with actual copyright, of course, as Google isn't the judge of that), and get taken down, and need to start a new channel from scratch. It's what we all get for using a free video streaming service to distribute our content.

I think the centralization plus trusting a company is the issue, not a lack of cost meaning we aren't deserving of a good experience.

Look at Peertube or other free as in freedom options.

> I think the centralization plus trusting a company is the issue,

I think that's not the issue at all. The problem is simply scale. Currently, about 5000 videos are uploaded to YT every minute. If just 0.1% of them have any potential copyright issues, that's 5 potentially complex cases per minute or 7200 per day.

No amount of human review will be able to decide this in a timely manner. The piano teacher's case is just the simplest of scenarios and you'd have to expect the vast majority being "fair use" cases, which are incredibly hard to decide.

Free an in freedom ultimately results in users being sued directly (see torrent networks) and I'm not at all convinced that that's any better.

7200 cases a day is most definitely not too much for a human review.

It is because people are routinely violating copyright in much larger amounts, yet nobody is bothering to go after them. Compare with the way the laws about audio and video cassettes ended up.

If they followed the DMCA instead of their own system it would scale just fine.

It would scale because under the DMCA system the site is not responsible for finding violations. That is up to the copyright owners.

All the site has to do is:

1. Take down alleged infringing content when someone claiming to be the copyright owner files a take-down notice.

2. Put the content back when whoever uploaded it files a counter-notice.

3. Tell the former that if they want the content taken down again, sue the latter. The site is now in the DMCA safe harbor.

All the site has to scale up to handle is dealing with notices and counter-notices. For that, all they need to do is check that all the required fields in the forms are filled out. This does not require anyone with any legal training--it is just checking things like they have identified themselves, described what content they want taken down/put back up, stated a reason for their belief that this action should be taken, and similar things.

You could train in an afternoon anyone who can read at a pre-high school level to handle this. 20 people in a normal shift could handle those 7200 cases per day.

Heck, you could even speed that up if you wanted by doing even less review on the notices. There aren't really any legal consequences to the site if they accept a notice that wasn't quite right. It is only the counter-notice that needs a little scrutiny. You want to make sure everything is correctly filled out in that, because it is the counter-notice that gives you the safe harbor.

I mean that's basically how it works right now:

https://support.google.com/youtube/answer/2807684?hl=en&ref_...

> > I think the centralization plus trusting a company is the issue,

> I think that's not the issue at all. The problem is simply scale.

Do you think that the scale of YT is achieved without centralization?

Given the reality that a lack of a centralizing force like YT would just shift copyright adjudication to actual courts and be more expensive and higher stakes for all involved and probably have a similar error rate and bad actors like Prenda, I’d agree that a relatively benignly uncaring Corp without access to police and prisons is better than the court route.

> 7200 per day

Let's say it takes 5 minutes to properly adjudicate a dispute. 5 minutes allows 20 per person-hour. An 8 hour shift could resolve 160 reviews. 7200 / 160 = 45 shifts per day to review all of the hypothetical copyright issues. I don't think that's required, and the number of requested reviews is going to be some fraction of that. Requiring google to spread out under 50 shifts over a 24 period in order to provide a fair review so that they may bring in billions of dollars a year from youtube doesn't seem like a large ask. This is call-center-esque work and even done in the US or Western Europe would be very cheap, particularly as it can avoid otherwise more cumbersome regulatory hurdles.

> Let's say it takes 5 minutes to properly adjudicate a dispute.

That's highly unrealistic, but review alone would take at least twice as long, since the average video length is about 12 minutes.

How would you find out in just 5 minutes whether a monetization claim is justified? Not every case is as clear as the piano teacher's case. Keep in mind that this isn't a DMCA takedown request either - it's about a party that claims the content in order to redirect the revenue.

So you seriously claim that on average you can find out in just 5 minutes whether one of the 6 license types [0] applies and the claimant actually has a case? If it was that easy, I doubt that court cases like [1] would take years. And that's assuming all the information is already at hand so no further communication with either party is required...

[0] https://www.bmi.com/licensing/entry/types_of_copyrights

[1] https://completemusicupdate.com/article/song-theft-dispute-o...

You don't need to review the whole video, the accusing side should provide the exact time at which the alleged violation happens.

> "fair use" cases, which are incredibly hard to decide.

Only a judge acting in a court of law can decide if something is fair use. All else is speculation.

1) Feed a computer with MIDI files of public domain music and render it as audio

2) Upload to YouTube

3) File copyright dispute to YouTube for any (future?) uploaded video which contains the music which used to be in the public domain

4) Have Google reject the videos

5) Create a site or an app which allows you to license that public domain music for a fee.

6) Notify YouTube who has licensed this public domain music.

Ok. that's the way Google thinks is the way it should work. Or maybe they just recognize that their AI is causing more harm than good.

Two days ago I uploaded a video which was a screen recording to demonstrate a bug in the Android app "Komoot". It was unlisted, the link was attached to the bug report I sent to the company. It was just a short video showing how the caching (or something in that direction) of uploaded images in the app seemed to be broken. The content in the video was 100% adhering to all the guidelines, specially to those of "for all ages". The content was nothing else but scrolling photos of an MTB-trail with a bit of UI. If your video is flagged or you mark it as "for 18+", then it can only be viewed by logged in persons.

After uploading the video I got an email that it was not complying with the "for all ages" requirements, which is kind of bad, because now the support team must log-in with a Google account to YouTube in order to see the harmless but useful video.

But then again, videos related to Instagram celebrities or Chinese ASMR-binge eating are totally ok for them.

> Ok. that's the way Google thinks is the way it should work. Or maybe they just recognize that their AI is causing more harm than good.

Let's be clear. It is not Google that wants it. Actually Google is on the side of calling all of this retarded. It is the state of copyright laws, of the DMCA, of the lobbying by the RIAA. The day the whole copyright ecosystem is updated to accept that computers exist and that people share files on internet easily, Youtube will be VERY happy to unplug all these terrible bots that are there to provide a bad solution to a problem we should not have.

> Actually Google is on the side of calling all of this retarded.

Google has the means to make this stop.

How so?

They just stop and wait for the DMCA take downs and copyright owners have to make their own claims? I guess.

Add friction to the process of submitting copyright claims. Reduce friction in the process of appealing copyright claims. Relax draconian rules like the copyright strike system leading to account closure. Adjust the content ID system to reduce false positives (assuming no magical improvement to the system this will come at the cost of increased false negatives; that seems like an entirely reasonable trade-off)

For starters, they should actually follow the DMCA. The DMCA gives affected people a defined way to counter-notice, and then the entity who filed the initial DMCA notice can either sue within 14 days or the content gets restored.

Instead Google chose to evaluate "disputes" themselves, with algorithms and "AI", and reject disputes. Rejecting counter-notices is absolutely NOT something the DMCA mandates or even suggests. Moreover Youtube essentially used to allow (probably still does) alleged rights owners to reject disputes in essentially one click, while the DMCA would require them to bring a law suit. Some copyright owners therefore created bots doing the clicking for them.

They could also be more lenient to "established" players as a first step, especially when it comes to counter-notices/disputes. Factor in previous history google has with an alleged infringer (alleged by their own algorithm by the way, not even by a third party) when considering a dispute, like account age, channel age, number of previous videos without problems, "we do know the customer" e.g. to pay out ad money, etc. And then maybe not outright reject it, but leave it to the alleged copyright owner to file a law suit (as the DMCA states) or at least refer it to actual human beings for further evaluation.

Of course, Google could hire people to check up on their own algorithms and decide on disputes instead of machines. Youtube had $6 BILLION in ad revenue in the last quarter (not year), so they could certainly afford to hire some people. In the end it might even be a profitable investment, as fewer good content is pruned from Youtube for wrong copyright issues, leading to more ad revenue.

Youtube right now seems pretty content with their quasi-monopoly, to their own detriment in my opinion. As unlikely as it may seem that people will create competitors, it can happen, ask mighty MySpace about it.

> Youtube had $6 BILLION in ad revenue in the last quarter (not year), so they could certainly afford to hire some people. In the end it might even be a profitable investment, as fewer good content is pruned from Youtube for wrong copyright issues, leading to more ad revenue.

Who is paying said ad revenue though? Could a large chunk of it come from the same companies/industries who currently enjoy the broken state of YouTube's DMCA process?

Parent comment was also referring to YouTube's AI marking a benign video as 18+. Google is willing to wrongly punish people via excessive/unfair false positives in the interest of trying to be more advertiser friendly. It's just about money.

Well, false positives result in fewer scandals than false negatives here.

So guilty until proven innocent seems like a perfectly reasonable, if very annoying, stance.

It's not reasonable to screw innocent people over.

It's reasonable for a for-profit company that has a monopoly.

No monopoly necessary to make this commercially reasonable stance to take.

(And, there's no monopoly situation here in any case.)

I don't think the laws are so harsh. It seems like YouTube caved in to the music industry interests so that popular artists continue to premiere their videos on YouTube.

And otherwise they'd get sued for hosting copyrighted content, which will result in hefty fines and jail( didn't the US want to get Kim Dotcom from NZ precisely for that?)

They won’t get sued if they follow the DMCA process which is very different from what they are doing.

They were sued [1]. The lawsuit lasted 7 years and ended in a settlement. The terms were not public, but I think it's likely they promised to institute a process that goes above and beyond what the DMCA requires. [1]: https://en.wikipedia.org/wiki/Viacom_International_Inc._v._Y....

So long as they respond to actual DMCA reports, no. ContentID goes far, far beyond what is required by the DMCA to have safe harbor.

Dotcom got in trouble because he was intentionally hosting infringing content, going so far as to try to hide that material from the copyright owners by making it look like he had taken it down when notified of its presence when in fact he just made the URL that the owners knew about stop working. Employees that went too far and actually took down infringing material got reprimanded.

Go dig up a copy of the indictment. It includes a bunch of internal emails from Dotcom and other running his site where they talk about all this stuff. It was basically a site whose intent and business model was hosting pirated movies. That you could also use it to host your own photos or whatever was there to try to provide cover.

Does the extent matter that much? Assuming YouTube did nothing to take down copyrighted content, is that better or worse compared to lying about taking down ?

The reason why Google is not accused of hosting copyrighted content is because they somehow managed to sell the fiction that streaming and download are two totally different things. The slightest touch can make that fiction dissipate.

Ehh, Google has gone to great lengths to ensure that the largest holders of copyright have their content libraries available to check against new uploads and apparently no lengths to ensure that public domain content is available for the same purpose.

The public domain doesn’t threaten to sue them.

It ought to.

Who has standing?

The public. It's ours.

Mskes sense, but how does that work?

Can any of us initiate a case and it becomes a class, or?

I'm open to having a conversation about it. What does the space look like between "no the public does not have standing", and "the public has standing individually or as a class"? What are the tradeoffs at those boundaries, and how can we work towards creative solutions that exist in the space between those extremes?

That convo should happen.

I dont believe this in the slightest. Besides, the false-positives which regularily pop up around YouTube claims are not a result of any law, or lobbying. They are the result of googles sloppy implementation of the cliam system.

https://www.youtube.com/watch?v=-eBB1ZzvFoI

There are disputes over "white noise" videos so it is already as bad as it gets.

Given that white noise is random, a recorded copy of white noise is its own perfect watermark.

Does it depend on seeding that would eventually have tendencies for replication? Or could that be pushed out until practically forever?

If the white noise is based on a random number generator, the pattern would restart after a full cycle.

If the random number generator's period is 2^32 and you use one integer per sample, then at 44 kHz, you would have about a million seconds, or twelve days, before the RNG has gone through a full period.

Most RNGs have periods much higher than this. xorshift128 has a period of 2^128-1 and the Mersenne Twister's period is 2^19937-1.

So you could push it out until practically forever.

So you could push it out until practically forever.

But, there will still be repeated sections.

Knowing nothing of the YouTube algorithm, I’d still wager it’ll take you longer than the lifetime of the copyright to produce conflicting noise works.

Depends on how lossy the compression is. Your watermark becomes compression artefacts instead.

Most audio codecs by now use perceptual noise substitution, which potentially substitutes in identical noise (I think it just EQs it?).

no, the mp3 codec actually deletes humanly inaudible portions of the frequency spectrum to reduce storage, this data isn't recoverd or replaced or referred when played back again, it's lost information. what is removed is based on human perception limits for audio

He is referring to modern codecs, not MP3. MP3 isn't widely used anymore.

Modern codecs don't encode noise - they remove it during decoding and then add back artificial "comfort noise" when decoding, e.g. for film grain or background noise in voice calls.

Got a source for the current usage of MP3?

What services do you know of that use MP3?

Virtually every podcast ever.

YouTube itself seems to rely more on webm for audio. That seems to be a container for Opus or Vorbis formats.

Vorbis and Opus themselves are lossy encoders (Ogg Vorbis).

> Virtually every podcast ever.

Apple Podcasts are AAC. Spotify are Ogg Vorbis. That probably covers 90% of the market right there.

The original question was whether or not audio could be fingerprinted from white noise, or if compression would obscure that fingerprint. The original context was YouTube videos specifically.

Sufficient to support the latter assertion was that lossy compression is frequently used.

YouTube's webm does in fact use lossy compression as noted. MP3 is, I assert, still significantly used in multiple contexts, including many podcast sites. I haven't specifically surveyed those, and don't use Apple Podcasts myself (never bought into the ecosystem, though as it happens, this specific response is coming from a MacOS system). I do make heavy use of tools such as youtube-dl and mpv, and find that those do in fact frequently find and extract mp3 audio from various sources, including podcasts and IIRC Soundcloud.

Verification is as simple as:

    youtube-dl -F <URL>

The '-F' flag will list available downloadable formats.

E.g.:

   $ youtube-dl -F 'https://soundcloud.com/danyork/tdyr-362-thoughts-on-wordpress'
   [soundcloud] danyork/tdyr-362-thoughts-on-wordpress: Downloading info JSON
   [soundcloud] None: Downloading webpage
   [soundcloud] None: Downloading webpage
   [soundcloud] danyork/tdyr-362-thoughts-on-wordpress: Downloading info JSON
   [soundcloud] 637919466: Downloading JSON metadata
   [soundcloud] 637919466: Downloading JSON metadata
   [soundcloud] 637919466: Downloading JSON metadata
   [info] Available formats for 637919466:
   format code   extension  resolution note
   hls_opus_64   opus       audio only audio@ 64k
   hls_mp3_128   mp3        audio only audio@128k
   http_mp3_128  mp3        audio only audio@128k (best)

Pedants are my fourth favourite people, but only during Lent on leap years, prior to Vespers, during a blue moon.

Re: opus

'Opus (audio format) - Wikipediahttps://en.wikipedia.org › wiki › Opus_(audio_format) CELT includes both spectral replication and noise generation, similar to AAC's SBR and PNS'

Which would mean that at least some of the white noise signature of an opus-encoded audio track would be ... a compression or encoding artefact.

Does opus have a notion of bitrates similar to mp3 or ogg vorbis? If so, the white noise would vary among bitrates.

Podcasts are pretty niche. Almost all codec use these days is from streaming services or video/phone calls - YouTube, Spotify, WhatsApp, etc. None of those use MP3.

What codecs do those applications use?

How do they preserve or modify white noise?

You haven't named any lossy stream codecs that do restore white noise, so i will not consider you a reliable source of info.

> Another notable addition in this version of the AAC standard is Perceptual Noise Substitution (PNS). In that regard, the AAC profiles (AAC-LC, AAC Main and AAC-SSR profiles) are combined with perceptual noise substitution and are defined in the MPEG-4 audio standard as Audio Object Types.

https://en.wikipedia.org/wiki/Advanced_Audio_Coding

If you take 30 seconds of white noise and look for it in 2h long generated output I expect that you can find similar 30 seconds in whole 2h.

Because if you use 30sec of someones song you have to pay up and youtube is enforcing that.

Now if you will take another 2h video cut it into pieces and start searching for similar patterns in other 2h video I expect you will find some matching ones.

If you're comparing sample-for-sample, maybe. But that's not what is performed by music identification algorithms - it would be both slow (to the point of being unusable) and inaccurate (because of re-encoding, etc). Instead they rely on different kinds of dimensionality reduction that match the way we perceive music, so they only compare smaller amounts of data to get more perceptually relevant matches. The Shazam fingerprinting algorithm is the best known but there are others.

So if you submit white noise to a copyright database, it could match different white noise recordings.

Even if they were the case, this further shows how limited and dumb (i.e. not 'smart') YouTube's copyright voilation detection system is.

You (probably) can't copyright white noise because US copyright requires authorship. So in the same way that you can't copyright a phonebook's alphabetical list of numbers, you can't copyright random numbers rendered as sound, unless you did something else unique to it to exert authorship. It's just not something protected by copyright law. So even copying someone else's exact white noise sample is probably just fine.

The problem is that YouTube's system can only apply simple content matching rules and it counts any sufficiently long content match as a violation with no consideration of the work or context of use. Thats not how copyright law works. Copyright is a complicated system with all sorts of issues like fair use, derivative works, public domain, and works that don't qualify for protection. It's not a database query.

I don't think it's hit YouTube, but there are people and/or bots enforcing a copyright on silence:

https://www.youredm.com/2015/11/28/soundcloud-finally-goes-t...

What would be worse: whoever holding the rights to John Cage’s “4’33” filing complaints against movies without background music, claiming that they have copyrighted background music.

> which is kind of bad, because now the support team must log-in with a Google account to YouTube in order to see the harmless but useful video.

Not only this, but in the EU* it requires verifying your age with a credit card or ID photo: https://i.imgur.com/gP20dXi.png

Since you van have a credit card when you are 13 yo, the check is more or less useless.

It's it effectively doxxing people under 13? That's a concerning implication if I'm understanding this correctly.

Honestly, the only way things will change is through grass roots abuse of the system like this. You have to force them to change.

Sounds like good material for a Defcon talk: "How I Own A Quarter of Content on YouTube"

You have YouTube's content rating system backwards.

It's not "racey content gets marked special" with "suitable for children" being a catch-all category.

YouTube's "content developed for children", is the special case. It's explicitly content meant for and marketed to children, or content that children would be particularly attracted to, like nature documentaries.

All other content should be marked "not intended for children", even if it's not "adult"--aka restricted to 18+--content.

This is stated pretty clearly in YouTube's documentation. They have a link to it in a contextual pop-up right next to the form field asking you to self-rate the video.

This was the email I got:

> We wanted to let you know that our team has reviewed your content and we don't think it's in line with our Community Guidelines. As a result, we've age-restricted the following content:

> Video: Komoot Bug

> We haven't applied a strike to your channel, and your content is still live for some users on YouTube. Keep reading for more details on what this means and steps you can take if you'd like to appeal this decision.

> What "age-restricted" means

> We age-restrict content when we don't think it's suitable for younger audiences. This means it will not be visible to users who are logged out, are under 18 years of age, or have Restricted Mode enabled. It also won't be eligible for ads. Learn more about age restrictions.

When I click on appeal I get a popup titled "Submit an appeal" with the body text of "Appealing this violation is not available"

----

EDIT: Ah. I see. I actually did set it to "Is made for kids" thinking that this means that through this option I express that it does have no content which would be against the community guidelines.

I've now changed it to "Not made for kids" but also "Not age restricted".

Thanks for pointing this out.

>or content that children would be particularly attracted to, like nature documentaries.

maybe times have changed, but you could not pay me to sit through a nature documentary back when I was 8.

nature content is a very weird gray area of content in terms of age ratings. It's technically just, well, nature. But then there can definitely be "just nature" content you don't want a kid to see.

Unless they've changed this you can get around this by sending the 'embedded' link: https://www.youtube.com/embed/<videoid> instead of https://www.youtube.com/watch?v=<videoid>.

http://spiderrobinson.com/melancholyelephants.html

TL;DR copyright becomes absurd surprisingly fast when you have a large population, widely-available authoring/recording tools, and a way to store/search all of them, indefinitely. Like, indefensible absurd.

That was a good read. Thanks for sharing it.

This is a fantastic story. Thank you for the link.

> Or maybe they just recognize that their AI is causing more harm than good.

Google recognizes that these flaws in their AI aren't worth caring about. Google doesn't have any mission or obligation to help the world share videos. Google cares about Google's profits. And they've found that the expedient way to do that is just let the AI be overzealous with rejecting, because the cost of a false positive is infinitesimally tiny and the cost of a false negative (real copyright violation) is so much higher.

How do we fix this? Competition. We need a Google/Youtube competitor so that users will choose the platform that does copyright recognition better.

What if someone exploited this to such an extent that Google could no longer ignore the issue? All someone needs to do is write an open source library that automates this. Almost like exposing a security flaw to the public to force the company to fix it. Sure it would enable bad actors but it would subvert them in the long run. I feel this may be the only way to force the company's hand as an individual.

A really great video (if you're willing to spend the better part of an hour watching it) about this is YouTube's Copyright System Isn't Broken. The World's Is.[1] by Tom Scott.

[1] https://www.youtube.com/watch?v=1Jwo5qc78QU

DMCA is bad but the undisputable claim and three strike systems is all of Google own doing

The video covers this aspect. Yes, the current system is bad, but it replaces a scary letter demanding cease and desists plus high damage payments and a resolution between extremely expensive lawyers. It surely is not optimal, but I doubt your common YouTuber could afford to fight claims in court.

That's a US-specific issue. Yet people all over the world are affected by ContentID (though at least in the EU AFAIK you can now get a third party to mediate this dispute).

DMCA replaces the scary letters. YouTube only responsibility is to collect claims and counterclaims. Common YouTubers can just accept the claim and mute content. Youtuber wanting a fight or that know the other party would never risk their frivolous challenge to go to court with the threat of perjury hanging on their head have a right to counterclaim and put the ball in the other party field.

YouTube instead decide to act as jury and will allow claimant to counter the counterclaim, and here lies the crux of the issue.

YouTube isn't protecting poor youtubers anymore than what DMCA already allows; YouTube is instead actively removing youtuber right to challenge the claim, putting all the power in the claimant, far above what DMCA mandates or requires.

And that decision to act as jury and judge and final unappealable authority in the claim process lies squarely on YouTube shoulders.

I don't think the DMCA replaces the scary letters. The scary letters existed because they were more effective than the DMCA. Threatening to sue someone is always and always has been totally an option. The reason scary letters don't get sent right now is because it's easier to just have Youtube remove the video.

If that went away the next easiest option (scary letters) would happen again. DMCA or not.

Even if that piece was copyright, is it a copyright violation to play it yourself and publish it?

Potentially, yes. There are different types of licences [0] and depending on your use of copyrighted material, it may require a license to publish it legally.

[0] https://www.bmi.com/licensing/entry/types_of_copyrights

Yes. Because music is both composed and performed, it has three primary copyright scenarios.

1. Print rights protect sheet music etc. from being copied.

2. Performance rights control whether your composed music can be performed.

3. Mechanical rights control whether your performance can be copied (onto mp3 say). Also related to these are "synchronization" issues and licenses.

Here are the “copyright owners” as displayed to the pianist in YouTube’s interface:

- APRA_CS

- ECAD_CS

- SOCAN

- VCPMC_CS

APRA is Australia & NZ, ECAD is Brazil, SOCAN is Canada, VCPMC is Vietnam.

APRA: Australasian Performing Right Association

ECAD: Escritório Central de Arrecadação e Distribuição

SOCAN: Society of Composers, Authors and Music Publishers of Canada

VCPMC: Vietnam Center for Protection of Music Copyright

I went digging. APRA = Australasian Performing Right Association Limited ECAD = Escritório Central de Arrecadação e Distribuição SOCAN = Society of Composers, Authors and Music Publishers of Canada VCPMC = Vietnam Center for Protection of Music Copyright

CS stands for collections society. In another age CS would stand for the Mafia.

All but ECAD were found from here: https://www.cisac.org/

These collection societies are actually the opposite of the mafia.

They collect royalties on behalf of the composers. If the composer has a publisher, the royalties are forwarded there instead (so the publisher can take their contractual cut).

They are the only way to protect your work if you are unsigned (think struggling artists).

AIUI there collection societies are collectives of publishers and look out for the people who own the publishing companies, they're not making sure the publishing companies act fairly and justly, they're making sure they sue people who are too small to defend themselves in order to boost the collections in order to support the publishing company owners.

Maybe I'm wrong ... counter evidence welcomed.

Something odd is happening here. Perhaps I can offer a piece of insight.

APRA is the Australian music copyright organisation (https://www.apraamcos.com.au/)

ECAD is the Brazillian version.

VCPMC is the Vietnamese equivalent.

Not sure about the others, but basically these are the people with whom a composer registers their work. These organisations work together by forwarding royalties collected within each territory to their rightful owners.

I am a member of APRA and very much doubt that someone got away with registering a Beethoven work as their own.

Very curious indeed. I have no explanation.

[Edit] You are only meant to register your work in one territory, so it is odd that this 'work' would be registered in four of them.

> I am a member of APRA and very much doubt that someone got away with registering a Beethoven work as their own.

I've yet to hear of one of these nationwide registries vetting any of the work that gets submitted to them for novelty or whatever. That would be too much like doing due diligence.

Strange, that nobody at Google seems to have heard of Beethoven.

IMO the biggest issue with all this is the lack of a suitable appeals process, handled by humans.

Would such a process be expensive? Yes, but you can't have it both ways, enabling copyright holders to lodge spurious claims at will, and not allow content creators - who the entire platform is built on! - to disclaim them.

Would such a process be expensive? Yes, of course - but YouTube can very well afford it.

Perhaps the copyright holder should have to pay for a human review, where the content creator agrees to pay the cost if it turns out to indeed be a true copyright infringement.

This would get rid of all bots and most false claims

That would be too much in favor of the little guy, I'm afraid, and probably doesn't align well with preexisting deals with the music industry.

If the little guy is indeed little and is earning little or none from their content, then is the copyright holder being harmed? Or is the copyrighted material being promoted, thereby increasing the potential value of that original material?

I think what should happen is this:

Step 1: Someone files a copyright claim (or perhaps this is done automatically). This is free.

Step 2: Someone disputes the claim. This is also free, and automatically restores the video.

Step 3: The filer can re-submit the claim but they have to post a bond for the price of a professional manual review by a trained copyright lawyer; maybe $1000.

Step 4: If the video owner can re-dispute the claim; to do so they also have to post a bond for the price of a professional manual review. The trained copyright lawyer comes to a conclusion; whoever "wins" the ruling gets their money back, and the other person pays for the whole thing.

If the video owner doesn't go on to step 4, obviously the copyright claimant gets their money back.

Would this not allow companies with deep pockets to continue to abuse the smaller channels who might not have $1000 lying around to put in escrow on the off chance of losing the $1000 just because some lawyer somewhere made a bad decision? Or they just get their video removed because they dont have $1000 to begin with.

Let's not let the best be the enemy of the good. Right now 1) companies with deep pockets can easily false claims with very little consequences 2) small channels basically have no recourse.

Obviously the whole system is predicated on the lawyers being fair: but assuming the lawyers usually DTRT, then 1) there will be consequences for false claims, resulting in far fewer of them and 2) small channels can get a real human to look at their claims.

No situation is perfect, but I'm pretty sure it would be better.

It would also create an arbitrage opportunity for investors who could inspect the copyright situation and put up the $1000 and collect at the end. The system would work well.

I'm pretty sure the process can't have payment stipulations like that.

That's sort of OK for an operating business, but many creators don't have $1000 sitting around for every video they upload.

I think the biggest issue is that the burden of proof and all the risk is on the person who's video is being claimed. False DMCA claims are free, practically riskless, and require no evidence.

> False DMCA claims are free, practically riskless, and require no evidence.

This is the crux of the issue. The DMCA needs to be amended so that filing an incorrect claim comes with some risk.

Copyright enforcement should be on the copyright holders. YouTube shouldn’t have to build AI to scan for copyrighted material. Copyright holders need to a person verify all copyright claims and be liable for the opposite exact same amount of damages that would be rewarded for legitimate copyright claim.

Talk to the RIAA etc.

They seem to believe that only their own sanctioned releases should be allowed, and that they should all be paywalled so that every view is paid for.

Or we could fight the whole damn robber baron system and return to something resembling common sense.

Actually, I have a better solution...

Disallow the uploading of all content held by RIAA/MPAA-affiliated companies from the likes of YouTube, even if the owners thereof wish it to be there. Fuck them. Let them go build their own platform if they think their work is so god damn awesome. Save YouTube/et al. for works created by, well... You.

>“copyright strike” on her account, wherein three such strikes would mean permanent closure of the account

I think the three copyright strikes have to take place within 90 days. Unless I am confusing community guidelines strikes with copyright strikes.

is worried about all the time she’s putting in to create content being usurped by a few bad actors

Is this where we must point out that the entire ContentID system was designed to protect bad actors, because it was written by them?

One could argue that the DMCA was largely written by bad actors. Everything else flows downhill from there.

The “western” world lost a once-in-a-thousand-years chance to change how copyright works at a fundamental level, when the internet started getting traction. We’re now forever beholden to the whims of parasitical “industries”.

There is actually one more step available, even after the copyright strike: you submit your contact details and YT clears the strike and it’s up to the claimant to then issue proceeding against you outside of YT.

https://support.google.com/youtube/answer/2807684

I have used it in the past and it works.

If it is so simple why are there content creators losing their channels over merit less claims?

I don’t know - are they submitting counter-claims? All I know is it worked for me and my channel although I was never at the point of having more than one strike.

> and YT clears the strike

there's the rub .. in the OP's example, the strike wasn't cleared

As I understand it from her video, she did not get a copyright strike yet, and she also did not complete the appeals process because she did not want to submit her contact details to strangers.

It seems that this entire fiasco can be fixed by simply having a media streaming platform which is hosted in a jurisdiction which doesn't recognize 'intellectual' property.

Why has no such thing emerged?

Such a jurisdiction does not exist. See Berne Convention, Rome Convention/PPPPBO, TRIPS, WCT, WPPT.

Well, presumably one will emerge at some point.

I'm just trying to look at the root problem here and how to overcome it.

ISIL? I guess when you have a culture that totally shuns the existing international framework they're unlikely to be able to or interested in competing with youtube.

Wow. So if a public domain piece of music (which does not have a copyright) is played by someone and recorded (which does have a copyright) the Google AI has no way to know if you're reproducing the the public domain music (which you are allowed to do) or reproducing the reproduction of someone playing a piece of public domain music (which you're not allowed to do).

I've already stated this many times, I'll state it again.

Get your shit out of YouTube and any other Google product.

Google is a dumb, faceless, fully automated company only interested in extracting as much data as possible from its users, force them to swallow as many ads as possible, all without caring about listening to them (both consumers and creators), under the faulty assumption that they're too big for users and consumers to live without them. They simply don't deserve anybody using their shitty products anymore.

The error in this case is quite obvious. YouTube's scanner incorrectly identified the teacher's recording of Moonlight Sonata as a copyrighted reinterpretation of the same piece of music originally written by a guy who actually died 200 years ago. And I can't completely put the blame on Google's AI: the notes are technically the same, the beat might also be the same, if you calculate an FFT of the audio you'll probably also come up with similar spectral signatures. But a human listener will IMMEDIATELY notice that was played by the teacher IS NOT the the same as the copyrighted piece of music.

The problem is: who is accountable for these mistakes? Who shall I reach out to if Google's foggy algorithms make a mistake? And, in the case of educators and creators who actually do that for a job, who will compensate them for the revenue they have lost because of algorithmic errors?

Until Google can provide an answer to these questions, I repeat: keep your ass away from anything that has their name on it. They are not reliable, the risk of losing your data, your account or your followers because of random automated decision is very high, and the probability of getting a real human to assist you is very low.

I agree with your assessment of Google; however, what is a reasonable alternative for a content creator who wants to publish their videos and be able to build an audience? You kinda have to go where the audience will be if you don't already have one and I'm not aware of any video discovery platforms with anywhere near the reach of YouTube. I've managed to get out of Google products almost entirely--YouTube remains the exception.

Moreover, let's say a new site comes along and dethrones YouTube. Remember that this whole mess started because of lawsuits that were ultimately ruled (or settled) in favor of copyright holders. Any player in this space will need a method for handling vast quantities of copyrighted material scanning and, like Google, will be heavily incentivized by legal precedent to have that system "err on the side of caution."

I'm not a fan of Google; but, the villain of this story is the horribly outdated and corporate-lobbied copyright system that will push any player in the video space to this kind of draconian approach.

I'd disagree slightly with the parent post here: the conclusion shouldn't be "get away from Google" so much as "If you have to use Google, understand what you're getting into." If you deal with unsavory people, things will go south eventually. If you deal with Google, sooner or later things like this will happen. Expect it, build it into your strategy, but don't be surprised by it.

How do you "build that into your strategy"? That is so easy to type, but unactionable. A piano teacher can't build her own streaming platform, and if she goes to a different platform she'll die in obscurity.

Concretely, what should she have done differently?

my thoughts would be:

- while building up your following on google, market your brand to your own site(s) as much as possible

- build brand an alternate streaming site(s)

- have a way to reach your followers in the event you get shut down - email list, etc. to inform your followers to switch to the alternate site

Do many people watch through embedded videos or do people watch in the YouTube app? There's a convenience in having all your videos in 1 place. I've had a couple podcasts move to Spotify only and I stopped listening even though I have a Spotify subscription.

It's that assumption of "dying in obscurity" that gives Google their damnable network effect.

If there isn't enough marketing advice out there about finding as many channels and methods as you can, I'll eat this comment with barbecue sauce.

Read up, utilize all the advice on everything except Google-owned properties, accept the lower visibility, and help neuter the algorithm beast.

Or accept that you're riding the tiger and never complain again.

> If there isn't enough marketing advice out there about finding as many channels and methods as you can, I'll eat this comment with barbecue sauce.

The advice is all there, all right. It's just that most of it is "join a platform that's NOT being used by literally everyone on the planet", which is not very effective when "literally everyone on your planet" is the "raw material" that you fish your viewers out of.

>Expect it, build it into your strategy, but don't be surprised by it.

This is kind of like saying... Expect failure, build failure into you strategy, and move blindly forward believing there is no other alternative.

I'd say it's more like:

Go ahead and build your initial business on quicksand.

Know that you'll get lots of traffic, but it might sink into the ground at any time.

Use the profits to build up a solid future property that isn't built on quicksand.

I'd say that accurately defines the entire entertainment spectrum in a nutshell. Content creator, game developer, asiring actor, etc.

Problem is few make enough money to get to your last step. And those that do can pay others to deal with the quicksand and are no longer interested in the fight to clean it up for good.

Explain this to my mom please. P50 users of the internet could not give two shits about these entities.

> what is a reasonable alternative for a content creator..?

Insurance? I'm only half kidding. No individual youtuber has the deep pockets to stir the slumbering Googlebeast enough to get it to notice and correct its mistake, but all youtubers certainly do. Conversely, perhaps there's a market for a we-only-get-paid-if-we-win lawyers to spring up here, as they have with workplace injuries and such. Youtube's resolution process may not be friendly to creators, but juries probably will be, if the creator bypasses Google and sues the party making the claim. "Sues for what?" I dunno - emotional damages? Tortious claims? They'll figure something out, I imagine.

Maybe those are silly ideas, but they're certainly less silly than waiting for Google to fix things...

Or join the youtube creators union https://fairtube.info/

> however, what is a reasonable alternative for a content creator who wants to publish their videos and be able to build an audience?

Put your videos on Youtube and PeerTube.

Does PeerTube have ads or monetization?

No. But you can advertise something in your own videos as people often do nowadays.

That is a lot harder to do, and only the most successful content creators will be able to find sponsors on their own. In addition, that is a lot of extra work for a small content creator.

... or the small content creator gets a random permanent ban for "REASON WE CAN'T TELL YOU SO YOU CAN'T GAME US", and the easy road turns out to be a dead end.

And you wind up doing the extra work anyway.

I'd honestly never heard of PeerTube so I searched it. Ah, they have a "about peertube" video! Should be great! In the first 30 seconds of the video, there were 1-3s long hiccups 4 times. Not exactly a confidence-inspiring start. After figuring out how to find more videos, I tried to play some--but, experienced more hiccups or load times from 20s to a full minute--and some just flat out didn't play at all.

I'm not saying PeerTube doesn't look like interesting tech; but this is clearly aimed at a far more tech-savvy crowd and to put it out there in response to asking for a, "reasonable alternative for a content creator who wants to publish their videos and be able to build an audience" is just totally missing the mark of what makes YouTube successful for creators and viewers alike.

> 1-3s long hiccups 4 times

PeerTube is not one single website. It is a decentralized platform, where everyone can set up their own server and it will work as a part of the whole system (like emails work). This is why it will never be owned by a single entity like Google.

You probably chose a slow server. It does not mean that the whole PeerTube is slow.

https://en.wikipedia.org/wiki/PeerTube

As stated above, I watched the promotional video from PeerTube themselves on the JoinPeertube.org website and experienced multiple hiccups in under 30s. If their own hosting isn't cutting it, what chance does anyone else have?

> You probably chose a slow server.

I didn't choose a server. I chose a video. The moment the service asks me to think about what server is hosting it is the moment I don't care enough to jump through those hoops. Never underestimate the value of a consistent experience.

Once again, I don't want to be disparaging to PeerTube--it's a cool concept and I understand the foundations behind it. But, spending 10 minutes with it earlier today made it obvious that it's not going to challenge YouTube as a content discovery platform.

Edit: Corrected the URL from "PeerTube.com" to "JoinPeertube.org" -- I typed "PeerTube.com" in haste and just assuming the URL.

> from PeerTube themselves on the PeerTube.com

PeerTube.com is not "PeerTube themselves". This is one of the servers, not the best one. This is the official PeerTube website: https://joinpeertube.org. It will show search results on many servers.

> The moment the service asks me to think about what server is hosting it is the moment I don't care enough to jump through those hoops.

You only choose your server once, like you chose Youtube once. You do not need to jump through hoops.

> If their own hosting isn't cutting it

PeerTube (actually FramaSoft) is a non-profit organization. You shouldn't expect huge resources from them. Also peertube.com is not their server AFAIK.

I corrected the URL in my post. I was on JoinPeertube.org and watched the, "What is PeerTube?" video that is embedded in that page. If I watched it again, it might not buffer at all. Maybe it was a bad moment. But I consume a lot of content on YouTube and I can't remember the last time a video buffered a single time--let alone several times in the first 30 seconds.

I'm honestly done with this ridiculous strawman about whether the video works. PeerTube is not a viable alternative to YouTube from a content creator's perspective for many reasons which I've already stated; but, I'll summarize:

1. It lacks even a tiny fraction of the distribution and discovery reach offered by YouTube.

2. It lacks the monetization features that allows YouTube to become part of a business.

3. It requires me to provide my own hosting and technical setup which is far more involved than dropping a video into your browser like you get with YouTube.

4. If I, as a consumer with no knowledge or interest in how PeerTube works, "choose the wrong server," I get a crappy experience with videos buffering for ages so I'm disinclined to continue to use the platform leading to reduced audiences on the platform and the feeling of, "doing extra work for nothing."

You can defend it all you want; but, your responses so far have been thinly veiled, "You're too stupid to get it right." I guess maybe I am; but, I'll stand behind that being the single biggest reason that PeerTube simply cannot be a platform to rival YouTube.

> 3. It requires me to provide my own hosting and technical setup

This is wrong. You choose someone's server and use it just like you use Youtube.

> "You're too stupid to get it right."

I never said or implied that. Yes, using PeerTube is slightly harder, but the benefit you get is huge. If it is not worth for you, you can give your live to Google...

> I'll stand behind that being the single biggest reason that PeerTube simply cannot be a platform to rival YouTube.

Well, I would count the greens first and then for rival, a joke. They can never share the love alphabet can share with the tube. You have to give money to take money.

The beautiful thing about technology is that anyone can do it. You feel like people are implying you're too stupid to do it, which no one will ever (or should ever) say since part of everyone's coming to terms with tech is screwing up in all the myriad ways required until you get it right.

Before YouTube, people self-hosted. That is still an option. You can drop your work product on a server, configure it to only accept requests from servers or locations you control, then link in your content til the cows come home. You may need a cache to handle higher traffic loads, but once you find a setup that works for you, you're golden.

Use Youtube for discovery all you want, but understand that someone else's computer will never be as immune to external sources of disruption and malicious bureaucracy as something you own and independently operate.

If it's important enough to become a revenue stream, it's important enough to accomodate some cap and op-ex to ensuring you have a fallback option for. The rest of the world can be relied upon, however, to want to get their licks in whenever somebody has nice things. Plan accordingly.

This is why we all can't get along and just have nice things.

>I can't remember the last time a video buffered a single time

happens quite often when trying to switch resolutions. Becuase Google wants to continually tell my 200 Mb/s connection that 480p is the optimal streaming solution, and now that is spreading to mobile as well.

> your responses so far have been thinly veiled, "You're too stupid to get it right."

I didn't get that impression at all, and I think we interpreted this thread very differently. The point wasn't that PeerTube would be the new Youtube, it would be that if you don't have terabytes of local storage to keep your videos, you can upload it to a place where DMCAs won't mean your videos being lost in the void.

No need for caustic language.

There are ways to avoid these copyright guys going after you with a truly decentralized system where node stores only a fraction of the content and there is no central place to ban content. Just like the concept of internet was DODs answer to a threat nuking communication there must be an answer to the threat of copyright trolling. I am very much waiting to a decentralized content platform to show up that is immune to content filtering abuse and much easier on the ISPs bandwidth wise than the current YT or similar sites.

> however, what is a reasonable alternative for a content creator who wants to publish their videos and be able to build an audience?

I think it's more instructive to look at it from the other side. What's the reasonable alternative for a video hosting service to doing this kind of policing? Remember it's not really an option to just throw video over the fence, DMCA requirements mean you have to be responsive. And thus there's a built-in incentive to cut a deal with the content owners to preemptively prevent the DMCA claims (which are expensive!) by doing this sort of automated policing.

It's true that not every host does this, but every host that doesn't do this either does it in violation of the law or eats significant overhead that needs to be recouped in some other way (i.e. by paying their content creators less! Check the author's channel, this is someone who's clearly on youtube for revenue. Would even she jump ship given that it would probably cost her money?)

Really, this isn't something Google can fix. It's a problem with the legal regime that imagines that all infringement is a bright line definition and that preemptive takedowns are the best solution.

The alternative is to responsibly scale your service. If your platform is so big that it can only be moderated by algorithms, and the algorithms don't work, then it's too big. Scale down until, at the very least, you can hire enough humans to review complaints when the algorithm does something wrong. Even better would be to have a human review every flagged violation to confirm it.

Obviously, hiring humans is expensive, and nobody is forcing them to do it, but that doesn't make it an unreasonable alternative. I consider it unreasonable to design an unethical system with the sole excuse that it makes more money that way.

So... which providers have responsibly scaled, in your opinion? All we have are tiny hosts (mostly porn) running by the seat of their pants and occasionally disappearing in a conflagration of lawsuits, and big folks like TikTok and Google and Facebook with draconian preemptive enforcement of various forms.

I think that argues strongly that the service you want to see is "unreasonable" given the regulation regime we have. You can't put this on the hosts, you'll just be disappointed. Call your representative.

To leave a ton of money on the table, and let the competition eat that space, because 0.1% of DMCA requests would be served with an overreaction?

Sorry, this is not going to make business sense. If you want a free video publishing platform that does user outreach for you, you got to pay a price; the false positives is a part of the price, alas.

That’s why we need a regulation change to make it so that you can’t outcompete by being unethical.

I don't think it's viable, but I welcome any attempts to actually imagine how such regulation might work, what would it consist of.

I definitely don’t have a general case solution. A solution for this case would be something like “a provider of video hosting services is not allowed to take down videos automatically without having a human review each request.”

The law never had that intent though, and getting it into there will most probably be impossible .

I’m not saying that we will get this outcome, just that it’s the outcome that we need. Unfortunately I agree with you that we won’t.

Easy. Put the burden of legal fees and risk of censure on a DMCA mill for false positives. This means if a host takes down content to maintain their safe harbor status, and the person can prove the copyright claim was false or frivolous, you get a cause of action to counter-sue the original claimant. Moonlight Sonata is in the public domain. The teacher's performance != anyone elses. Therefore, false, therefore it should have reprecussions for the signing attorney. If it was pre-emptive by Google without a DMCA request from an external entity, that's a different issue.

People don't get it. Perjury actually means something. When you have the blade of perjury over head, it is absolutely the case that as a human being, if you have doubt, you should be saying it, or you're misrepresenting the truth of the matter.

The level of perjury inherent to generating these claims via automated process is absurd. It is absolutely reasonable that when you have a group going around and using the legal system as a cudgel, the proper response is to return the favor.

This process should not scale at all if people would punch back. Perjury should be trivial to prove when no one even looked at the content in qustion aside from an analysis suite.

Regardless; there is always the self-hosted option.

I agree it doesn't make business sense, but it does make moral sense. And yes, this hypothetical service probably wouldn't be free.

well what kind of a monstrosity is this buisness sense then when it is strcturally incompatible with any moral whatsoever (as seems to be more or less the consensus of the first couple comments here)

Completely agreed. I'd love to see Google (and let's get Twitch in there as well, while we're at it) working with content creators to push for legal change. To me, that's what's most disappointing: Google will spend millions on lobbying to legally collect more and more information about me; but, they don't have the inclination to use their size and scale to push a significant expansion of Fair Use Doctrine (or, if they do, they certainly aren't vocal about it).

It was out of self interest of course, but, Google did just finish a decade long, costing probably hundreds of millions of dollars in legal fees, battle with Oracle to fight for reasonable copyright interpretation for software developers.

This current situation on YT is the result of another multi-billion dollar legal fight with the music industry. I don't like how YT handles this either but I put most of the blame on the music industry for it.

Also remember Google Books, which was a huge (seriously, huge) project that spent years with full funding only to be killed by the publishing industry because they couldn't justify it under the notion of "fair use" we're stuck with.

you are getting to things confused here: fair use of copyrighted material, and being responsible and responsive for a quasi judical algorithmic process and its mistakes.

(remember its beethovens moonlight sonata, one of the most famous classical pieces period. i dont think you even could copyright a part of it, at least not just the sequence of notes)

> And thus there's a built-in incentive to cut a deal with the content owners to preemptively prevent the DMCA claims (which are expensive!) by doing this sort of automated policing.

How are DMCAs expensive?

You need to pay a human being at $15/hour or more to spend a few minutes to review the claim and response, on videos that on average are probably making you a dime or less in ad revenue. These companies receive a lot of DMCA claims, most of which are absolutely valid.

>I think it's more instructive to look at it from the other side.

I'm cynical, so I feel it's much more likely that a competitor will push actual change to come (be it through google self improving, or through future lawsuits challenging the current laws) than for a current monopolistic entity to finding alternatives to problems that don't inconvenience them at large.

>Remember it's not really an option to just throw video over the fence, DMCA requirements mean you have to be responsive

"Responsive" is the key word to be challenged here. I'm unsure if automating a removal at the behest of any barely or unverified account is the bare minimum "reponsiveness" required legally. The big problem that won't be resolved without someone legally challenging it is that there's no negative consequence to filing a DMCA claim. Or at least, there wasn't as recently as 4 years ago.

> what is a reasonable alternative for a content creator who wants to publish their videos and be able to build an audience?

I've noticed several tech-related content producers copying their videos over to LBRY/Odysee as a backup in case the YT algorithm decides to cancel them.

These are great solutions to keep your content if Google decides to lock you out or destroy your content; but, they aren't platforms on which you can build a content creation side of a business.

Yeah - agreed.

If I was a creator I’d use YouTube, but also have backed up local copies of videos. Preferably hosted elsewhere in case the channel gets in trouble too.

You have to be where the users are and YouTube is by far the best video streaming service (with the largest audience).

Everyone uses youtube. Local businesses use youtube. I took driving lessons and they use it. Small tech companies use it to host their promo videos, and so on. So people use it without even needing the "reach" that youtube can provide - it's just being used for hosting.

Again and again the same basic "but it's free" refrain.

Sure, building on quicksand is free. No one wants to pay for a service that is as undependable, and impossible to appeal in case of fraudulent or just mistaken bans.

So go ahead and build on that quicksand... but do it knowing so, and stop complaining when the quicksand swallows you.

There might not be much land around the edge of the quicksand patch, but it's solid ground.

I think the point here is that users don't have much option, not the hosts. If the DMV uploads instructional videos to Youtube, it's hard to boycott google if you want a license.

Meanwhile, I don't think the State govt.is going to care much about convoluted processes against takedowns. That's just Tuesday for them, they aren't gonna fight google on this despite technically having the resources.

All that is true of course but I think the bigger issue is that the copyright system has to go or at least change dramatically. People walk around with video cameras now 24/7, they want to add music to their video. We should facilitate this.

More and more, ordinary people are not citizens in a nation, with a balance of rights and responsibilities afforded to them, but simply cattle: a resource to be cultivated, controlled, and harvested. This is true online, but also increasingly off. People are still operating under the outdated notion that corporations, banks and government institutions can be held to account because they are in some way vulnerable to the displeasure of ordinary people. They cannot, and are not.

Youtube does not care about your happiness. Their business model is unrelated to it. Youtube, like Google, is in the business of gathering up a bunch of delicious users and bolt-gunning and butchering them so they can be served up to ad partners for a tidy sum. In this arrangement, the happiness of the livestock hardly matters.

>Get your shit out of YouTube and any other Google product.

My thoughts exactly 5 years ago. So I did exactly as you say back then. So , did you notice? Did it help? I do not think so.

May be the problem is more complicated and leaving YouTube would not help ?

I guess it did help you, since you are protected from Google's unpredictable moves.

>you are protected from Google's unpredictable moves.

Am I? It seems I am still influenced by their moves. Of course doing nothing is a good protection against any failure except perhaps a total failure to deliver anything useful? It is predictable, yes but I am not sure we can count it as protected result. It is more like protected nothing

> under the faulty assumption that they're too big for users and consumers to live without them

It is, unfortunately, not a faulty assumption. I'm a piano teacher on YouTube, and I'm able to make money there. My entire audience was developed through the platform. I put videos on platforms like Vimeo or even Peertube, and I've had more views in one day on YouTube than their entire lifetime on those other platforms combined.

The network effect is a cruel mistress, but short of some kind of global exodus no one has the ability to change that. I hate it as much as anyone.

Any creator with established audience can bring those people to alternative platform, e.g., Peertube. Start from posting videos on both platforms and advertise the other one. Then post videos on Peertube earlier.

That will only work for a select group of viewers. Waiting a day or even a week isn’t a huge deal for me as a consumer, and I’m familiar with it because Patreon people often put their videos up in patreon earlier than their YouTube ones.

You are right. However, when Google blocks the creator, most of their audience will know how to find them.

My understanding gained from watching various kickstarters and patreons is that creators don't need a million views, they need a few thousand engaged fans and they can be self sufficient at that point.

The question, always, is what's the alternative? Rumble?

Yeah, but YouTube has ads that can pay the content creator. PeerTube doesn't.

Any creator with established audience

Well yeah, once you're already successful you have more options. That's absolutely no help to people who do not have a paying audience.

This is simply incorrect. I have seen many creators and businesses try this and simply not find movement, for example RoosterTeeth which owns many different channels and brands and has a very comprehensive video streaming site, yet their videos get 1/1000th the viewership on their own platform vs youtube. The platform is captive.

In all fairness, I think Mixer employing this exact strategy showed that it is extremely hard to migrate communities. Even with creators making millions of dollars a year.

As a content creator, you want the platform to protect your copyright and at YouTube's scale, copyright scanners are the only way to implement this.

Will they produce false positives? Of course. Do the benefits to content creators outweigh the costs of these false positives? Yes if you believe creative content should be protected.

The only other viable model for content creators are subscription based services like Patreon and they have/will also be pressured by the entertainment industry or even the content creators themselves to flag copyright infringement once the platform gets large enough.

As it stands, someone can make a remix of your song and make claims against anyone using your original song. It all happens through youtube's automated system, and you won't see a penny for your work.

> As a content creator, you want the platform to protect your copyright and at YouTube's scale, copyright scanners are the only way to implement this.

Is this really true, though?

Google certainly doesn’t want to pay for people, but automated systems can be used as a first-pass filter, before human are brought-in to make a second-pass judgement.

So what is the rate of false-positives? And how many automated flags are triggered per day?

Like some have suggested, the automated system should only be an input for a team of humans to then review and consider. They should then contact the potentially breaching party and get some feedback as well as doing some due diligence to check if the complaint came from a proper right holder.

Yes this costs money but it is the only way to do the job without being a scumbag and a general burden on the world.

Furthermore it should cost the person sending a complaint somethings to file it, which then is refunded if the complaint is found to have merit.

Lastly, some of these steps could and should be skipped if a particular account or multiple accounts determined by some other means to be the same person are repeatedly found to be in violation.

Similarly if a person keeps making unfounded accusations the refundable fee might increase in steps.

That's all great in theory, but current US copyright law isn't really in favor of implementing any of that. If Google tried to do it, they'd likely get flooded with massive lawsuits from the media conglomerates because they'd lose the DMCA safe harbor.

No doubt you’re right. In that case the copyright laws are wrong and should be fixed.

I apologise and transfer my scumbag stamp from google to Congress then for this particular case :)

Why would they lose the DMCA safe harbor? We're talking about YouTube's own scanner here, not DMCA takedowns. There's no requirement in the DMCA to implement an internal scanner that has many false positives, and not investigate the scanner's output.

But the difference between composition and performance is well delineated in copyright law. What technical barrier is preventing Google from setting up scanners to detect similarity, get a hit, and then identify a video as a performance of 'Moonlight Sonata', look that up, and OK it on the basis that everything written by Beethoven is long out of copyright?

>Will they produce false positives? Of course. Do the benefits to content creators outweigh the costs of these false positives? Yes if you believe creative content should be protected.

False positives are absolutely unacceptable when it comes to anything dealing money. I sure wouldn't stick with a bank if they said "oh well, false positives happen. you'll get your money next week... maybe".

but there are several banks that help ensure that security. There are no checks to google allowing for such false positives.

Sobhow was wicca moonlight even copyri if it was that close to something that is out of copyright protection?

People are using this system intentionally as a scam. Recently, I tried uploading a video which had video game music in the background, and youtube flagged a copyright claim on behalf of someone who made a remix of the original song. I did not have a license to use the song, so that's fair, but my choices were either to drop my video or run ads that give money to someone who wasn't even the owner of the song I was trying to use.

There was no avenue for recourse, or to report the person who was fraudulently making claims.

Can you reach out to the original composer of the game music? They would likely care about the bogus claims. Particularly if it is a smaller studio. It’s not a great solution, but it might work.

Im not excusing what happened, but was the remix author actually making claims? Or did they just add a song to ContentID’s database? If it’s the former, that’s despicable and (almost certainly) illegal. If it’s the latter, it’s not their fault, but Google’s for having an overzealous system.

Without a viable and better alternative this wouldnt make sense. But what would be a viable alternative? Well, it could use only non-copyrighted material, which is feasible nowadays, there is tons of CC0 music to dress any video

> The problem is: who is accountable for these mistakes? Who shall I reach out to if Google's foggy algorithms make a mistake? And, in the case of educators and creators who actually do that for a job, who will compensate them for the revenue they have lost because of algorithmic errors?

Given that we've had to endure this nonsense for more years than I can count now, and Google and other companies refuse to budge, it sounds like a great opportunity for legislation to enumerate users' rights in situations like this, and to enumerate penalties should companies like Google violate them.

And put your content where? What should this piano teacher do?

This is partially the fault of YouTube/Google, but also copyright law in general. It's broken.

This community, right here, could collaborate and force changes upon the powers that be. Boycotting Google and YouTube isn't going to do it though.

> And put your content where?

Vimeo? Peertube? Facebook? Self hosted website? Anywhere?

You can put your content anywhere you want but you can't choose where the audience is. YouTube is in practice a big monopoly because there's no audience in the "alternatives." If you want to be seen you have no choice but YouTube, you can prepare for the worst and upload simultaneously to other platforms and include links in your video description but you can't avoid YouTube.

>Vimeo? Peertube? Facebook? Self hosted website? Anywhere?

Those proposed alternatives don't address why the content creators like this piano teacher put their tutorials on Youtube:

++ $0 in hosting and bandwidth costs: self-hosted costs money that's often unpredictable, and Vimeo has platform membership fees

++ ad revenue to help make the effort of producing a video worthwhile : Peertube does not have relationship with ad sponsors

++ audience size & reach : Vimeo/Peertube/selfhosted/etc don't have comparable viewers. For niche content such as piano instruction, this makes building a financially sustainable audience more difficult

++ discovery recommendations from the platform: Vimeo/Peertube/selfhosted don't have the network effect ecosystem of other videos on music that can lead viewers to the piano teacher's tutorial videos.

When frustrated Youtubers ask "And put your content where?", they're not looking for dumb hosting sites to upload some mp4 files. Their question is really a short version of: "And put your content where that has the audience reach and monetization to make the video production worthwhile?"

A content creator like this piano teacher wants to make some extra money with Youtube videos. It's not the end of the world if she can't do that but the extra income could help offset the cancellation of in-person lessons because of pandemic social distancing. I don't think lecturing people repeatedly about Vimeo and Peertube is helpful.

EDIT reply to: >But the other rely to this comment makes a very good point - post in as many places as possible/desired _in addition_ to YouTube, and point to all the other places in that YouTube posting.

You're still losing sight of this thread's topic: the piano teacher is losing ad monetization money to a fraudulent claim of copyright. If she hypothetically uploaded her Moonlight Sonata tutorial to Peertube/Vimeo/selfhosting, she still gets $0 in ad share revenue from those alternative video hosters which makes the advice irrelevant.

Your "syndication" advice to distribute the videos to multiple sites solves a different problem such as de-platforming. E.g. Youtube deletes/censors her video or her entire channel.

That's not the problem she has. Her video is still there and viewable. But she doesn't want the ad monetization money stolen from her by a fraudulent claim.

But the other rely to this comment makes a very good point - post in as many places as possible/desired _in addition_ to YouTube, and point to all the other places in that YouTube posting.

Also a good way to inform viewers about Google's shitty policies and forewarn them that the much more reliable sources are All The Others.

>But she doesn't want the ad monetization money stolen from her by a fraudulent claim.

she's making $0 now regardless. May as well do a small jab in spite of it by promoting other places. More likely she'd probably just stop hosting videos and switch to a different format, like private tutoring.

Facebook? I hope you're kidding

I hate copyright law and agree it's broken, but it's not at fault here. Compositions can be copyrighted, but obviously Beethoven's Moonlight Sonata is in the public domain and has been for a long time. Performances can be copyrighted, but obviously a performance isn't the same thing as a composition.

There's no legal grey area here. This is completely on Google/Youtube. Why isn't there a way to assert the copyright status when uploading, beyond saying you ownt he copyright or not? The answer would seem to be that it would take some work on the company's part, and they don't want to put it in because it's unlikely to yield any additional revenue and the occasional bit of bad publicity doesn't hurt them enough.

"This is completely on Google/Youtube. "

What about the copyright trolls that are filing claims for content they don't own, forcing Google into an arms race with them.

Even when the system is working "correctly" it is broken. If I take a video of my kid dancing to a song, I often can't share it with my family on YouTube. That's messed up.

Is it messed up, though?

Suppose your video goes crazy viral, has 12 million views, you monetize it, earn some modest figure on the thing. What forces you to pay the song's creators/copyright holders? What would make you think you should not pay them?

Obviously, this is an unlikely scenario, but where's the threshold at which it becomes important? Monetization? First million views? Theoretical monetization? Actual payout?

that notion is why I imagine Google added a threshold before a channel can be monetized. So the payment details here are moot unless you are already an established channel who should know this. Regardless, this is less about a random user "making money off music they don't own" and more about the potential that users aren't instead going to Vevo to make the music company ad revenue.

I have my qualms on DMCA as a whole, but this is ultimately tangential to the real issue of the lack of consequences for perjury.

Google's extension to copyright and lack of accountability is the only thing broken here. That, and the fact that they have multiple monopolies.

Boycotting Google is the only way to go. One can also advocate for regulation in this are.

As for your question: They can put the video in Nebula or Patreon, for instance. Maybe there can be more of those, perhaps for music teachers. Maybe that's also an opportunity for someone new to jump into the streaming game and provide some competition.

Many, many creators who's content is not youtube friendly are hosting their videos off platform (no shortage of options there) and only posting previews on youtube to direct their subscribers and anyone who happens upon their videos to an external site.

Google makes this moderately difficult of course, but it seems to be a viable option when coupled with patreon (where you usually get more content/early access/more creator engagement) or some other external subscription where the same is offered.

To answer your question succinctly, no, you don't want "other youtube" you want something sufficiently different to avoid youtube's pitfalls.

>To answer your question succinctly, no, you don't want "other youtube" you want something sufficiently different to avoid youtube's pitfalls.

I argue "other youtube would suffice". We see this with Youtube streaming needing to actually compete due to Twitch existing, and vice versa. There's still the likely issue that they both fall into legal troll pitfalls, but I argue that they will avoid more of them in an attempt to compete for a better service.

With video hosting, Youtube doesn't need to compete.

I think you are referring to porn. Yes, there are places people can post porn and get views and monetize them.

I don't think that helps piano teachers and the like.

Interestingly this intersects with https://news.ycombinator.com/item?id=26996972

Content cannot be safe if the tech necessary to deliver it is out of reach for independent creators.

Note that in the Roku thread Google is (IMHO) the good guy, for reasons unrelated to this thread's concern.

Cannot be safe? What does that even mean?

When I was at Google they had a mission statement that went something like "Organize the worlds data and make it available." But the reality seems to be it was "Seize the world's data and hold it for ransom." The latter of course being the better business model.

In the case of YT, it's not clear there's much "seizing" going on. YT "users" are uploading petabytes even without being asked to do so. Not clear what the ransom letter says, either.

The business model is closer to "use network effects and really well implemented streaming technology to build a monopoly on video hosting, and then use the insane level of viewership to extend our advertising business"

See that is the 'hook' then Google monetizes those petabytes without paying the people who uploaded them[1], that is the 'business model'.

[1] https://www.cnbc.com/2020/11/19/youtube-will-put-ads-on-non-...

I still don't think that this constitutes "seizing". YT provides the streaming service for the uploaded video, claims the ad revenue in exchange. I don't like it, but it's not "seizing".

Also, boring as it is, and aware as I am that I am in a tiny minority, I do not use a mobile device, and on my browser(s) I have uBlock Origin installed, as well as 69k entries in /etc/hosts to block any and all access to a huge number of known-to-serve-ad domains. I essentially never see YT ads.

> Get your shit out of YouTube and any other Google product.

Any video hosting service that you will move to will behave in a very similar ways, in order to avoid getting sued out of existence by the media lobby.

If you want to solve this problem, the solution can only come from dismantling the parasitic parts of the media industry. That requires political action.

Sure, that's step 2.

Step 1 is walk away from the biggest, most obscure and capricious banning system around, to pull the biggest teeth of that copyright industry.

If you think that's step 2, then all you'll be doing is playing whack-a-mole as the MPAA/RIAA will either sue hosts out of existence, or strong-arm them into implementing something functionally identical, or worse than ContentID.

The MPAA/RIAA aren't stupid, and they can smell money. Hosting firms have money. Users uploading infringing content to a service gives them a stick to beat the hosting firm until money falls out.

I just want to emphasize that Google are not the only ones,

I had my Sony Play Station account banned for 2 months with no exact reason and no way to appeal, on short the Sony message was like "you did something wrong, something about our policy on sex and violence, our moderators are perfect so there is no mistake and there is nothing you can do".

I think we need something that addresses all such issues and not a Google only workaround, the solution is regulations.

Genuine question, what should I replace Firebase with? I have a responsibility to my customers there, not just myself. I want to fulfill that responsibility by migrating off of Google while continuing to provide the product performance they expect.

The error in this case is quite obvious...

That was the symptom. The cause of the error was Google allowing another Publisher to stake claims on public domain music.

The more stories like this crop up, the more I wonder what the source is. Google's pervasively horrible treatment of both free users and paying customers reflects its extraordinary fetish for using computers to do a person's job. The worse it gets the more I start to think this is coming from someone in particular. How could they be so incapable of improvement unless someone powerful is stomping on it? Who is it? Who in the top twenty Alphabet/Google executives has this godawful obsession?

There is no viable alternative and there won't be without changing the law. And the law won't be changed due to lobbying (any success would be temporary)

You can escape Google's algorithms, but you can't escape these sorts of dumb algorithms generally. People get all sorts of bogus DMCA notices, spit out by some application that scours the web.

Beethoven's copyright has long expired, and all music of the 18th and 19th centuries is in the public domain. (Disney notwithstanding)

Along the same lines, I was publishing church services each Sunday between May and December of last year, with three English hymns in each, mostly old ones well-known across denominations. In total, I got 23 Content ID claims (about 20% of the hymns), claiming ownership of the melody. Every single one was for a work in the public domain. I disputed each, providing the name and date of death of the composer.

Two were released (one after three days, the other after 27), the rest were allowed to lapse in my favour (which happens after 30 days).

A few of these claims were duplicates (same claimer, same hymn). This showed that even though I had successfully disputed their claim in the past in the grounds of the work being in the public domain, YouTube had not revoked their ability to claim the melody in question.

If any of the liars claiming they owned these works had rejected my dispute, it could have become a strike, &c. and there would of course be no recourse, because Google generally refuses to arbitrate.

Also, an insidious aspect of the claims system was that YouTube basically didn’t tell you about the copyright claim at all when publishing; you had to close the edit page for the video and return to the list, and see if the Restrictions column for the video said “Copyright claim”. If you didn’t do that, the video would probably be being monetised by the claimer in at least some of the world (and if you disputed it, I guess Google would happily take the lot for the period while they had monetised it—though now that they’re putting ads on everything, this lot isn’t so different from the usual situation; depends on whether you hate the balance of the ad money going to the copyright liars or Google more).

(There was also one amusing case where some music being played at a nearby temple was audible during a quiet time in the service, and so I found out the name of the music being played. I claimed fair use on that one, because I didn’t even want that music in it, and it was quiet. That claim was released after ten days.)

I don't think I'm ever going to approve of any sort of automated copyright claim system but if Google wanted to make it one bit fair, they should use the same concept of three strikes against the accounts that make false claims and ban them.

Too easy for bad actors to create multiple accounts or game this system. I think this would end up hurting genuine claimants of actual pirated works more that it would help the wrongly accused.

If you're a genuine claimant, you have little to worry about because you don't make false claims.

That's putting a lot of trust in YouTube that they'll find every claim you make to be valid. Someone could make a dozen accounts with a dozen duplicate/similar versions of your song, you dispute all of them, are found in the right on 9 claims and in the wrong on 3, your account gets 3 wrong accusation strikes and you channel gets banned while the other person then marks one of those knockoffs as the official one.

> Every single one was for a work in the public domain.

While the sheet music is, the performance by other artists - as your own performance - is not. That's the issue the algorithm is having here (not defending).

The automated system would need to "understand" that this is indeed a new performance of a public domain piece of sheet music and not a reproduction of a copyrighted performance by somebody else. Even if you were note for note playing exactly the same (tempo and whatnot) with the same instrument tuned the same way. I think this would be an argument against automated systems. Whether a human could know that my bike-ride was scored by myself and not somebody else is doubtful though.

On the other hand, I do understand that people do not want their individual performances to be used without licence and there may be many such performances.

Every time that I’m speaking of it was the melody that was being claimed, not a specific recording.

I understand that that is what was claimed.

I said that two actual and copyrightable performances of the same melody are arguably indistinguishable.

Edit: How would an automated system know, that you did not just non-transformatively alter another person's performance, instead of performing yourself?

I have gained the impression that YouTube has two distinct techniques within Content ID, one for claiming melodies, and another for claiming specific recordings.

I’m absolutely confident that this is doing melody matching: these recordings are trivially distinguishable from any professional performance, with completely different instruments and playing styles, and with much lower quality singing.

For example, the most repeated claim was by “AdRev Publishing”, claiming “Crimond (The Lord's My Shepherd) - FirstCom” three times. (That song was also claimed once by “Capitol CMG Publishing and Adorando Brazil” as “The Lord's My Shepherd I'll Not Want”.) The first two times, I was accompanying with a piano-sound keyboard in the traditional four-part harmony—admittedly they sound fairly similar to one another; but in the third, an Indian was playing, using a piano-and-strings sound on a different keyboard, using Indian harmonisation (which is quite different).

I think we even had an unaccompanied song claimed once, matching throwaway0b1’s report.

These are completely different performances from whatever recordings the liars may have provided to the Content ID system. The melody is the only thing they will have in common.

My experience aligns with yours. I had a video of a half-hour coffee-shop-style gig blocked because at one point I covered "Hotel California". This was just me -- one voice, one acoustic guitar played poorly, and was down a whole step from the original key.

I believe that Hotel California is still under protection as a melody, rather than a performance. Hence covers require either a license or fair use according to law. This differs a lot from the case of performing a 200 year old song whose melody is in the public domain.

Not to say it was wrong what you did, or that Google was in the right for flagging. But Google was legally correct.

(Google suggests controversy around the question of whether hotel California is public domain or not)

> How would an automated system know, that you did not just non-transformatively alter another person's performance, instead of performing yourself?

Yes, exactly. You have really nailed the problem with the current system.

Google chose to build an automated system. That was a choice, not an immutable fact. Google chose not to have a human arbitration process. Another choice. Google chooses not to punish the claimants that abuse the system. Another choice, not an immutable fact.

People aren't saying there must be a perfect automated system. People are saying Google has chosen to employ an automated system that does not meet the actual needs.

The criticism is for Google choosing to exclusively use an automated system, which can never successfully perform this task.

Ideally, the easy cases are automated, and more difficult cases like these are not. Hence, the system would know because of an actual DMCA request, rather than an automated YouTube request.

I've had the same experience (although I don't remember if it was specifically the melody being claimed). I don't always bother to dispute them (in my understanding, most of them just claim ad revenue, and we don't run ads), but sometimes it annoys me enough that I do. (The organist and people singing are pretty clearly pictured, and I get especially annoyed when it's a capella.)

> we don't run ads

Historically Google only put ads when requested by the channel, which required a fairly significant threshold of views and subscribers and supposedly manual review by Google. Some time last year they started a switch towards serving ads on all videos, regardless of the preferences of the channel (whether you’re big or small, whether you want ads or not), which I hear has been progressing steadily further and further. (I wouldn’t know. The internet’s too dangerous to view without an ad blocker. I also just generally hate ads and only see any at all when I leave my peaceful rural environs and go to the big city.)

True, this is also something I should look into further (also, for example, to make sure that they don't force ads when a copyright claim is put on).

Of course, that I don't ever look at it without an ad blocker makes it somewhat more difficult.

My reading of the situation at the time was that if there was a copyright claim, the video would be monetised in certain regions (depending on the claimant). So once I cottoned on to the situation, I always disputed the claims before publishing, so that no one would be fed ads. Now, who knows. Once I returned from India to Australia I stopped uploading the videos personally, and I don’t intend to publish any of my own stuff that I might make to YouTube.

Yeah, I should probably just dispute them all at some point.

This is fraud. They claim they have right to something they don't, when you dispute, YouTube just directing them the dispute and they reject.

As you don't want to risk 3 strike, you agree to share revenue.

Basically Youtube doing nothing.

Have you ever tried to report fraud to authorities? Even if you show overwhelming evidence, details of perpetrator, chances are LE will laugh you out and say it is a civil matter.

With law enforcement at least you get to talk to people. With YouTube/Google all your complaints go straight to /dev/null

Only if you're a big enough company and/or have to the right contacts. Do you think regular people could get the police to raid a data centre or something on grounds of copyright infringement?

no, but a 10YO troll in Brazil sure can get the FBI to raid the home of an American streamer they don't like.

Not saying it's even by any means, but it's better than zero human response. There will at least be some case in a govt. record should some bad actor keep getting reported, unlike Google.

LE might also decide they look funny and launch a frivolous prosecution against you.

At least DMCA has a perjury clause. It seems like there is no punishment for spamming frivolous takedown notices on YT.

> Basically Youtube doing nothing.

Actually, they're hurting an independent content creator over something she has every right to do using works in the public domain. Youtube is doing that, not the fraudster making the false claim. They have no power other than what Youtube gives them, and Youtube gives them 100% of the power.

I can't understand why Youtube has no procedure to flag companies as "Wicca Moonlight” like scammer / troll so, after a certain number of infractions like that one ( pretending to be the rightful owner of the rights on BEETHOVEN music !), they are sanctioned in some way or banned. That to add some symmetry to the procedure to fight trolls trying to steal other peoples hard job.

Maybe "Wicca Moonlight" has previously been flagged as a scammer/troll and are now on their tenth account they've been doing this with? It could be whack-a-mole on all ends. It's not particularly hard to spin up another LLC, register, and start filing claims.

They might have such a system, but the scale is so big, that we are still seeing the false negatives?

It seems like the claimers have little invested into the persuit compared to the people making videos and could relatively easily make a new scam account. It certainly wouldn't hurt to add a little more friction to someone making frivolous claims as it would likely kill off a number of them that are low effort.

Alphabet gets paid either way.

I think the problem is that the whole system they have is to appease copyright holders so that they don’t get sued for individual videos. They’ve productized it and made it all look very official, but the underlying fact that this is about appeasement puts them on a bad footing for banning trolls. If they ban someone, they’re basically saying “sue us instead”, which is exactly what they don’t want.

"Wikka Moonlight" seems to be the name of the audio the copyright claim is originating from rather than the organisation claiming the copyright. [1] The audio contains a recording of Moonlight Sonata with a load of added reverb. The uploading channel is "Alice Violet Molland - Topic".

The "Topic" bit generally gets added to a channel name when a music distributer (like DistroKid [2]) publishes music to youtube on behalf of a musician (in this case Alive Violet Molland), typically at the same time adding it to other streaming services.

Distrokid will let you publish an unlimited number of albums in this way for about $20 a year and then collects the revenue from any streaming on your behalf. I'm guessing the distributer may also register the audio with music rights organisations, which are presumably the source of the copyright claims.

The Wicca Moonlight video now has 4.4k downvotes.

[1] https://www.youtube.com/watch?v=-ThyLB6bakk [2] https://distrokid.com/

> The uploading channel is "Alice Violet Molland - Topic"

I went searching around using that info and came across a (similarly outraged) thread on google's support forums about this video, in which someone named longzijun seemed to make a reasonably sound counter-argument:

https://support.google.com/youtube/thread/108213944/a-posted...

> Then she files a counter-notification. The appeals process has not been completed. She is only part-way through it.

> Once she files that and if she does it properly, the claimant has 14 days to initiate a court action against her or the claim is released, the strike removed and the video goes back online.

> Obviously, the claimant will not pursue legal action in this case.

> False claims can cause inconvenience for sure, but with the counter-notification system, they don't do long-term damage.

> Both the takedown system and the counternotification system are mandated by US law (specifically the DMCA).

> YouTube is not supposed to intervene in copyright cases. If they do so, they will lose their safe harbor status (again under the DMCA) that protects YouTube from being sued for hosting copyright infringing content.

> To sum up

> 1) the dispute process has not been completed in this case

> 2) your beef should be with US legislators, not YouTube

Issuing the strike when the process is not complete is an example of them getting involved.

Especially since there’s no strike for copyright holders if they issue a false claim.

Unless I’ve misunderstood something, no DMCA notice or counter-notice has been sent.

Instead the copyright claim is purely via YouTube’s content ID system that detected the supposedly infringing audio before the video was even published, as the YouTuber states in the video linked. The company claiming the copyright sought to monetise the “infringing” video for themselves through this content ID mechanism.

I would argue that, while YouTube says it cannot arbitrate copyright disputes, if it continues to allow supposed copyright holders the exclusive right to decide whether something is in fact their copyright, they are arbitrating the disputes, just in a completely one sided manner, and the creators beef should be with them. If they were actually neutral, they would instead allow the DMCA system to work as you said.

That poster may be right on this particular issue... But it's hard to take them seriously after umpteen posts where they go on yammering about "If someone posts a recording of someone's performance" after repeatedly being told that the lady posted her own performance, and repeatedly asking someone else whether they are that poster, which AFAICS has fuck-all to do with anything.

So much agree. Their taking issue with that detail made me trust their viewpoint less.

How is it the fault of that youtube channel? They may genuinely want to protect their reverb'd rendition of the Moonlight Sonata, which is technically their right even if it's stupid.

Then it's purely google's fault that their AI contentid scanner can't distinguish between Wicca Moonlight and some other arbitrary performance of the sonata. It's google's fault that their AI doesn't understand that the sonata itself is in the public domain so they have to only match against exact reproductions rather than "kinda sounds the same because same notes and instruments" reproductions.

There are non-expired mechanical reproduction copyrights for Beethoven music. (Basically any time someone plays from Beethoven's score, or a modernized but public domain version of it a new creative work is created with the corresponding copyrights.)

So for this reason YT might have some minimal justification for handling "old classical music" (but it is not based on copyright of a long dead composer, but on the "mechanical reproduction" copyright of some orchestra/artist).

Now, that said. If they can't distinguish between performances, then it should not be left to them.

And, anyway, the real problem is the fucked up bias in favoring the claimers.

I used the YouTube studio editor to add a piece from YouTube's library of music in the public domain to a small video of no real consequence, and then, ten years later, received a copyright claim on that audio. The system is really truly fucked.

I feel like the fundamental issue here with these giant tech companies is their scale has outgrown the capabilities of their AI. They've all basically bowed down to the mantra of "AI will solve everything", and indeed they are so big that getting human customer support at the scale they need would be a huge financial hit, even for them.

But here is a situation where any rational human can say "this is crazy, anyone with sense knows that Moonlight Sonata is in the public domain", but Google has created a Kafkaesque nightmare because (as has been lamented a million times on HN) it takes an act of God (or Twitter outrage) to actually connect with a human support person is you're not a paying customer.

This entire episode is a great example of how the SV crowd really is full of groupthink and ideology despite their veneer of being rational. Everyone who isn't in the SV RDF can see the limits of AI today for example youtube copyright strikes, failures of AVs to take over the streets, and so on and so on. AI is useful but it clearly has limitations but the industry can't recognize it because they're high on their own supply.

the ai is built to deal with the peasant class. the ai occassionaly abuses them, because the peasants have no power. but nobody cares about the peasants except other peasants. the peasants are fed enough (you tube mostly works for them) just enough to keep them content and any sort of uprising is quelled. content enough in their toil (creating content), the ai grows stronger.

what a world to live in

This sort of thing happens way too often on Youtube, and it's far worse than merely violating copyright; here artists get denied ownership of their own expression of public domain music. Unlike mere copyright violation, this is actual copyright theft that Youtube is enabling here.

It should be punished harder than merely copying someone else's work usually is, but instead this sort of direct theft seems to be allowed by governments and copyright institutes.

> here artists get denied ownership of their own expression of public domain music

Hell why stop there, music artists get their own original music stolen by Youtube who then proceeds to hand it to someone else, for anyone interested in a popular example: https://www.youtube.com/watch?v=z4AeoAWGJBw

YouTube's automatic system exists to keep the company from getting sued and it's going to work like anything else at Google and be as automated as possible. The necessary result is a black box ML system to score content, and biasing ambiguous output in favor of the side more likely to have lawyers on retainer.

She actually decided to quit YouTube after this completely unfair treatment.

I think it's high time these platforms saw some regulation. It's ridiculous what they are allowed to get away with.

Good for her, it's the sustained highly creative output of skilled people that makes Youtube interesting beyond music piracy and memes. They have the power to collectively do the same to any other platform and shouldn't tolerate abuse like this.

You know that Google's behaviour here is a reaction to regulation?

It's tempting to answer failures in regulation with cries for more regulation..

Hardly, dmca is straight forward law, take down notice and counter notice has no three strikes. There are penalties for incorrect claims to deter such claims

This is Google's poorly implemented system to satisfy big studio creators to publish on YouTube.

This seems more a failure in Google's reaction than in the legislation they're reacting to.

Quitting will just get people to use the next best teacher on youtube that they can find. She can just get the music muted and create a link to her patreon, and tell that Google muted it. Google will lose out on revenue that way for now.

Actually I doubt that you can avoid abusing current copyright laws.

Also because there is no such "right" in nature. If you wish monopoly call it for what it is and have laws in accordance. Such "right" shouldn't exist in the first place or at least in the current form.

Copy process is the way people learn and progress. Copy process is essential for spreading knowledge and development.

When you learn you copy. When you sing you copy. When you speak you copy. When you teach you copy. When you think you copy. People copy their parents. You are the mixed copy of your parents plus mixed copy of other things and factors.

When you show to others what you did you copy too.

I give you a concrete example. I was dancing Argentinian tango in a charity event and wanted to show record of it to other people. I've got copyright claim on YouTube.

I was not even demonstrating music - I was demonstrating Dancing! The music played in the background was created somewhere in 1930-1940. The music played at the event was not even original composition. It was a cover done by local artist. And it was played live during the event with improvisations. Unfortunately I couldn't show this event to anyone else because I've got copyright claim on YouTube channel. I couldn't even share the record with my friends as private link. I couldn't even show the video to my partner for goodness sake. If this is not idiocy resulting from so called "copyright" then what is?

Unlimited copy process is crucial to arts especially because you should feel free to express yourself and only this way something new can appear.

Copy is a natural process and any attempt to regulate it or regulate it too much can and will create more problems than it solves!

Perhaps it's time to start listening RMS more carefully. This is one about copyright: https://www.youtube.com/watch?v=eginMQBWII4

Better terms for copyright include "imaginary property" and "bourgeoise pseudocapital."

Not only does copyright hinder creativity and infringe upon your natural right to copy, but it doesn't even benefit artists so much as companies who pay them a pittance while wielding the law as a cudgel to protect their imaginary fiefdom. The world needs less ownership, not more.

People will complain about copyright precisely until they have some creative work of theirs sold by someone else. Professional authors cannot exist without copyright.

I complain about copyright precisely because I cannot even show my creative work. And if I'll try to upload it anywhere where people are then I will be stripped off any of my "rights" in that very moment. So how copyright helps in that case?

Your problem wasn't an issue with copyright but with YouTube's approach to handling copyright.

Copyright is a way to encourage the production of new culture. Even if it is abused it doesn't remove all of its merits. Countries with weak copyright laws do produce or export content and are barely heard of and the reason why is not hard to deduce: all their top talents can't live out of what they produce and therefore go elsewhere.

If you really want to go the deconstruction route, laws are unnatural but murder and assault are found in nature everywhere and therefore we're all doing it wrong. Now if you excuse me I have to take care of my collection of scalps /s.

>Copyright is a way to encourage the production of new culture.

Perhaps it just was? Things evolved. Digital tools now make everyone face the law that was invented for publishers while not every person is even aware about publishing nor wish to deal with it.

People simply try to express themselves and punished for that like they are criminals. I do not think this is a way to encourage the production of new culture today.

From encouraging tool in the past as it perhaps was today it become a tool for opression of any form of self-expression because the environment has changed. Now it does exactly the opposite of what it suppose to do.

>Now if you excuse me I have to take care of my collection of scalps While you take care of your collection of scalps please notice how valuable such activity was back then and how useless it is now.

You basically use one single anecdote while YouTube has literally made more millionaires among independent content creator than mainstream TV did in all its existence. The reason they can produce the content they do is that they can take have copyright protecting their work. The notion that copyright is "a tool for oppression of any form of self-expression" is false just on its face.

And if it weren't for for copyright most of programmers wouldn't have a decent salary and open source projects would have less protection.

There are certainly reforms needed in IP, but abolishing it completely is just throwing the baby out with the bathwater.

You basically use one single anecdote while YouTube ...

Here is another anedcdote from one of the famous YouTube creators:

I Confronted the People That BLOCKED My Video (Rant) https://www.youtube.com/watch?v=RqvuEal2P2E

I Testified Before the SENATE On BLOCKERS | Here's What Happened https://www.youtube.com/watch?v=qhxLuFx-Sbc

This is really bad! YouTube should find ways to respond to misclaimed tracks as a creator. Also companies / individuals that claim public domain music should be banned from the platform.

you're fundamentally misunderstanding the purpose of the copyright claim system: it doesn't exist to determine the legitimate owner of a work, it exists to keep youtube from being sued. Since there is no legal penalty for a "false positive", even if the creator can prove that it's a false positive, while there is a penalty for a false negative, the system will always lean to aggressive removal and demonetization, because that's where youtube's interests lie.

the legal answer here is that if you don't like youtube's copyright system then distribute your content on another platform. Otherwise, push for there to be a penalty for an excessive removal of copyright, because that's the only thing that's going to tilt youtube's scales back to the center.

> Since there is no legal penalty for a "false positive",

I’m really confused by that part: YouTube lawyers must know that there’s a significant risk they’ll piss of a someone like a law professor and that they’ll use all their faculty to go after what is essentially fraud: there are provisions in the CDMA precisely for that, and the platform is liable for indulging those in systematic cases. Even if it’s not, they are risking a change in law.

Compared that PR nightmare to having an engineer spend a few days to hack a “public domain” user that automatically accepts re-use, or re-train to distinguish interpretations from copies… I feel like I’m missing something — and I’m not missing a cynical take on how Google is too big to care.

God knows that I’ve been arguing that some problems are harder than one expects, but public domain sheet music isn’t a hard one for YouTube.

The nuance with "public domain sheet music" is that while the instructions of how to play that music is public domain, the recorded performance of such music is not. So when you play moonlight sonata, you have the copyright to that particular performance (recording) and someone else cannot use that without permission. It is somewhat harder to distinguish between recordings of thousands of performances of the same musical piece.

The thing though is that youtube decided not to care. Actually, that is not about caring - the youtube we knew when it first launched was killed because it was an impossible business model if they themselves had the liability for uploaded possibly copyrighted content. So now they say "sort it out between yourselves, go to courts if you want but I will take it down while you do that to not be liable in the worst case".

This particular case is about a copyrighted "melody" and not the performance. Sheet music would cover this.

The same principle applies though: Youtube has no say in deciding what is copyrighted or not. They don't want to have a say in it themselves. Someone, somewhere, claims that they own the copyright to x and y. Youtube can't and won't decide if that is truthful or not. Almost a decade ago, youtube gave the keys to claims to people to remove themselves from the equation. "Someone says they own this stuff so they took it down. Sort it out between yourselves."

How much would you bet that the same wouldn't happen to e.g. Warner Bros.. YT is quite definitely taking a stance, the problem is they're not taking a stance on truthfulness, but simply on how much money does it make us.

They are taking a stance on how much headache and money it will cost them, which is really reasonable. The reason old and unsustainable youtube died is because of the Warner Bros and alike - because their kind tend to hold an enormous amount of copyrights. If they feel they benefit from copyrights, then they will feel the most attacked. They, with their influence, said youtube cannot continue in this fashion back in the day, and it was true. They could sue youtube to the ground and they'd have to close down (assuming manual moderation is impossible). I am actually amazed that they allowed it to continue with this solution, being, "youtube, you give us the keys and we do the takedowns when we feel like our copyrights are violated, then you may continue" and youtube said "fine" and that is where we are. Now if you feel like your content is taken down unjustly, you should take it up with whoever is claiming ownership over your content. Which is... impractical actually, but that is how it is. This is one of the only ways within the current legal framework where you can have a site where people post content to freely without being liable for aiding and abetting copyright infringement yourself. No easy solution to this.

So yes, Warner Bros can be an enormous headache to youtube. Legally and financially. "Piano teacher" probably can't. Youtube acts accordingly which is pretty rational. If you owned a youtube-like service, you'd have to do the same.

I see where you’re coming from. But a class-action suit from nearly every user that has used public domain copyrighted material and couldn’t monetize it seems just as bad.

> there are provisions in the CDMA precisely for that, and the platform is liable for indulging those in systematic cases.

(assuming you mean DMCA) Yes, that has provisions, but YT copyright claims are explicitly not DMCA claims. This is intentional to both protect the content creator (lawsuits are expensive!) and YouTube.

> YouTube lawyers must know that there’s a significant risk they’ll piss of a someone like a law professor and that they’ll use all their faculty to go after what is essentially fraud:

YouTube can ban you from the platform at any time for any reason. While it's not good publicity, you don't have a legal right to be on YouTube or receive money from them.

> Even if it’s not, they are risking a change in law.

I doubt Google is happy with the status quo, actually, since it is not good for their creators. But with the system being as it is, they're doing what's best for them.

> Compared that PR nightmare to having an engineer spend a few days to hack a “public domain” user that automatically accepts re-use, or re-train to distinguish interpretations from copies… I feel like I’m missing something

For one, recognizing this music is not an easy problem. You'll need to be accurate in a wide variety of cases, lengths and qualities, which in itself is already very hard. But, just because it's that piece, it's not necessarily free: Recordings of these songs can be copy-righted. For example, Beethoven's music itself is free, but the recording performed by the Sydney opera is not. So you need to decide whether the uploader has rights to this specific recording, which is nearly impossible.

And this is just the easy case. Fair use, for example, is a very gray area and something which can take courts years to decide. Same on whether a piece is derivative or different enough to be its own work. There is no chance for Google to automate away any of this.

> There is no chance for Google to automate away any of this.

I'm not quite so pessimistic. There is a chance, machine learning is pretty good these days.

But: the chance is pretty low and I assume other priorities are taking up most of their time, and this would be a risky project from a legal point of view.

There is no chance because copyright is not a data problem. It is a provenance problem. Provenance is not a function of data.

(Humans can't really solve this problem either.)

> Provenance is not a function of data.

Might not be a function of the data in the video, but they can throw more data from elsewhere at it.

>YouTube can ban you from the platform at any time for any reason. While it's not good publicity, you don't have a legal right to be on YouTube or receive money from them.

no, but being compliant in fraud goes farther than arbitrarily banning a user. Not a lawyer, and odds are there's not enough care to address this point legally anyway. But I don't think the potential case here is as open and shut as "we have the rights to refuse service".

Well they already managed to piss off the EU, but after intense lobbying by Google, it looks like that the resulting law hinders more YouTube's competitors?

Sure, YouTube doesn't want to get sued, but they also don't want to lose most of their content creators. Things are rarely as simple as assuming a party has a singular motivation.

Or creators should promote public domain platforms like archive.org instead of complaining that a commercial platform is run like that

And what exactly would that solve? Now you've shifted the problem to a different organisation without addressing the actual issue: bad actors and copyright trolls being asymmetrically favoured over content creators.

Being a non-profit organisation doesn't make you immune from frivolous lawsuits...

> And what exactly would that solve?

Google won't get the ad money from these views. That's the point.

This has been happening for decades. Almost all musicians playing classical music has had bans, 3 strikes, blocks, etc on social media, youtube, live streaming sites.

Often times Hilary Hahn (one of the great violinists of our times) gets copyright claims on her own practice streams because they sound literally like her concerts and solo albums.

Basically, the more refined and the better you are as a classical musician, the more likely you'll get a ban for streaming live music (that you play) because ... the composers are all long dead and there's thousands, if not tens of thousands of previous interpretations of the same music (literally note by note the same) floating around in the algorithms of the RIAA ban bots.

Rick Beato explains a lot of this very well. He's even testified in congress about this subject. Here's a good explanatory video of his on this subject: https://www.youtube.com/watch?v=E5lY_DbUsok

It's slightly different as the music he uses actually has existing copyright, but along the same lines.

"You can't teach people music without giving examples. I can't do something like, 'well you know, this is like Led Zeppelin, or like Bach'...this is something I created and you're going to learn from it... No, you have to learn from the actual sources."

Being able to access real people for support should be a mandatory requirement for big corporations offering services to the public directly.

In the UK you have a right to have automated decisions on e.g. loan applications reviewed by a human. Broadening a right like that would certainly be worth considering.

The GDPR has exactly this provision.

https://gdpr-info.eu/art-22-gdpr/

No. Mandating real people won't help. I've dealt with real support people who are exactly as useless as the website's FAQ. They're powerless to make any special exceptions to the predefined process or even to escalate to someone who can.

As an example, I can't use my main email address for an Apple ID account because apparently somebody else set it as their backup email and I may have carelessly clicked the accept link when I got the confirmation. I talked to a human Apple support person and his higher level colleague and was told that's probably what happened but they can't know for sure and even if they did, they can't fix it. The end. Bye.

You could try doing an account recovery (leverage that email) and just change it to something else.

I think I lost the email which was many years prior, and also I would need to know the actual email of the account it's linked to, which I don't.

at least you spoke to someone and now know why your screwed vs an automated AI system that doesn't tell you anything /s

To be fair one of the major drivers of quack medicine is that there are still a lot of problems where science based medicine will eventually go "We don't know what's wrong with you, but it's probably not something we can fix, so it just sucks to be you right now."

It may be the best most honest answer, but a lot of people would rather have a name for what's wrong with them, even if there's no cure, and a treatment, even if it doesn't work.

>Mandating real people won't help.

Mandating real people COULD help, bust sometimes it doesn't - yet this is by design. For example, Amazon Seller Support renders humans into bots because they can only reply with templates (it's like they have humans teaching machines what to reply from a fixed set of replies) - of course this is a shitshow.

If you get a cryptic reply, you have to figure it out, just to reply and get the same response, and then to finally get a "case is closed".

At least it's better than nothing.

> Being able to access real people for support should be a mandatory requirement.

Being able to appeal an automated decision to a real person is a mandatory requirement under the GDPR.

https://ec.europa.eu/info/law/law-topic/data-protection/refo...

Especially noteworthy here: > The data protection law establishes that you have the right not to be subject to a decision based solely on automated means, if the decision produces legal effects concerning you or significantly affects you in a similar way.

That seems to me (IANAL etc) like it could be used to argue that you need to be able to appeal to a real person who actually has the power to do something about the problem. Because these examples like in the GP(?) where one gets hold of a real person who claims they can't do anything because "that's just how the system works"... Well, then they're not really "a person" in the sense that I'm fairly sure has to be the one meant here; they're just another cog in the automated means.

I didn't know this. Thank you. I hope FB finally allows me to appeal against the block they have placed on my domain on sharing debugger citing "community guidelines".

Or in this case, maybe we should be relaxing the legal requirements that lead to YouTube trying to enforce copyright.

While that would be good, I don't think this would solve anything on Youtube. In the end Youtube is heavily courting the music companies with the way they do things, it's not about the actual law.

See the whole world of educational music youtubers, they are well within their rights to do quite a few things, but in reality they can't even perform certain short guitar riffs themselves without getting flagged.

I don't understand why YouTube doesn't use their content id system to detect these obvious public domain disputes.

Probably because it would cost them time and effort, but they have no incentive to do so. Why would they? At most, when cases such as this come to the public's attention, all that YouTube suffers is inconsequential reputational damage. This is easily and cheaply mitigated in PR terms if it gets bad enough by just handling the incident with a quick "Oops, sorry, our bad" and reinstating the removed material.

Meanwhile, the uncountable cases of illegitimate copyright takedowns which never make the news don't matter. What are the victims going to do about it? Sue? Use a YouTube competitor?

> YouTube suffers is inconsequential reputational damage

No such thing exists. Using "reputation damage" in context of YouTube, or any Google service really doesn't make any sense.

Even the marketing you mentioned isn't really necessary. That's what happened when you're a monopoly, bad reputation doesn't really hurt the bottom line.

The only defender of the rights holders without money would be the court system, so a court needs to suspend youtube's copyright strike system.

I don't know what I find more annoying; the ridiculous US copyright system that allows people to copyright almost 'anything'... or the unfathomably stupid people employed by the likes of Amazon, YouTube et al whose policies seem to be "Ban first. Ask questions later... Bu let's not bother with the questions"

It doesn't surprise me that a 200 years dead composer's works are considered subject to copyright when, apparently, even referring to the name of a 1200 year old mediaeval manuscript violates copyright:

https://stiobhart.net/2021-04-8-bubble-trouble/

The policies are directly caused by the laws. If google does not react swiftly, it is considered guilty of copyright infringement. Remove the laws, the policies are instantly removes as well.

Exactly. Benefiting from Safe Harbor provisions is usually predicated on taking the claimant at their word and implementing restrictions as fast as possible.

No such requirements exist on the release side.

Therefore it is no surprise that platform policies will heavily favor claimants, that is very strong incentivised if not an explicit requirement.

  >If google does not react swiftly, it is considered guilty of copyright infringement.

Only if a copyright infringement has actually taken place.

The problem is that none of these companies ever apply a gramme of logic or examine the merits of the claim, when someone cries 'copyright infringement'. They just automatically remove the 'offending' article and refuse to countenance any counter-arguments from the person accused.

It's this supine attitude which, I reckon, is fuelling all these ridiculous claims. I just wish that the likes of Amazon, Google, YouTube, RedBubble... etc. would call the copyright trolls' bluff occasionally and not just instantly cave. Every. Single. Time.

> The problem is that none of these companies ever apply a gramme of logic or examine the merits of the claim

True, but, TBF, evaluating merits doesn’t scale, the authors of the safe harbor provision knew this, and this is exactly the outcome the law intended, though it does not mandate it.

I studied there. When I got to the start of the long winded official name, I actually said "oh no" out loud to myself :(

Youtubers have it coming. They have accepted obviously unacceptable terms of service. The problems of Youtube are very old, and people have been consistently proposing alternatives that solve these problems. Yet they willingly chose to fatten the beast in the hopes that they will get spared of its cruelty. I say good riddance!

There should be some kind of law to forbid this type of unilateral model training. What great lapse of judgement to not also train the same model that is used to blacklist music to also whitelist music that is in the public domain -- and to use AI in this way, to suppress the creative freedoms of billions of human beings as the robot revolution edges closer to automate us out of existence. AI should not be used this way, but if you're going to do it you may as well be honest about it. How this algorithm could be deployed without first even acknowledging the existence of music outside of what is proprietary is disturbing

What a horribly broken platform.

Thankfully Peertube appears to be gaining some traction due to its compatibility with Mastodon.

Yes, it expect it will have a near monopoly in the year desktop Linux goes mainstream…

Funny but I can actually see peertube appealing to big YouTubers, especially if there's an ecosystem around it, like apps and fully hosted servers.

Really? Interesting as i am looking for a video hosting platform.

Looking at their front page and it looks dead and weird to me. A mix of my little pony videos, scantily clad women and 5 second long videos of grass, bugs etc. Absurd curation.

I used to have a bunch of time lapses on YouTube- my office for a few years was above the ferry docks in Seattle. I chose music from the YouTube collection of free music for background music. I have copyright claims on all of them now; I’ve taken almost all of them down. https://youtu.be/TgzqOvTvWzw As an example

btw I loved your video and I would prefer to see them even without music. I can play guitar while watching or play some other music in background.

thanks - I appreciate that.

I believe this is the video/music that claims the copyright of Moonlight Sonata (notice the downvotes): https://www.youtube.com/watch?v=-ThyLB6bakk

The 2015 album is on Spotify too:

https://open.spotify.com/album/1eMRlKHhUiVQRncGQRBBmH?si=K3e...

Just seems like the artist played a slowed down Moonlight Sonata with some effects on top and the algorithm now thinks it's theirs. Hard to know if they know that they are blocking/leaching from other content creators.

it makes it worse that its a hamfisted, soulless rendition. perhaps Mx Molland has other superior tracks but this one is particularly egregious.

I can't be certain, but I think this video might be the "original" that the algorithm matched: https://youtu.be/-ThyLB6bakk

What is to stop bad actors pumping out a stream of interpretations of public domain melodies to prime the matching algorithm then using the robotic nature of the appeal process to extort ad revenue? Nothing, apparently!

Susan Wojcicki on the case again. The freedom of expression pillar would crumble without her support.

I made a Moonlight Sonata cover myself a few years ago (https://www.youtube.com/watch?v=gIbBcJ7EkxQ) and had countless copyright claims since then.

I had to manually dispute every single claims over the years and prove again and again that no copyright were infringed.

It can get tiring when you get a copyright claim finally lifted after weeks of dispute, only to get a new claim the next month.

I wonder if small claims court could assist, for each infraction and claim. Time lost, etc.

IANAL

For music copyright is split in 2 types: composition and recording. The copyright on the composition is probably expired, but on a recording most likely isn't. The automated system most likely claims it's a copy of a copyrighted recording. Since it's a popular piece, it can be hard to tell recordings apart.

Note that the composition might also still have a copyright because it's not the original composition but a derived work, but that seems unlikely in this case.

One of the first uploads I did to YouTube, a recording of me playing BWV 1034-3 (Andante) got a copyright claim. This was the day I realized youtube is in fact helping big corporations to supress individuals. "We might delete your account if you raise an objection against this claim and we find your objection is invalid." That was pretty much around the time YT decided that it can now be as evil as it wants.

This kind of absurd copyright trouble is all over the place in the world of music education. I advised my boss to choose a video hosting platform between:

- a paid video upload account in a place where we're not the product (vimeo)

- self hosting (PeerTube)

YouTube's Copyright System Isn't Broken. The World's Is. https://www.youtube.com/watch?v=1Jwo5qc78QU

As much I love tom scott, he is ignoring the choices that YouTube made along the way were that precisely that choices. They didn't have to become this kind of platform.

Tom himself is owning/promoting nebula an alternative platform with quality curated content.

Youtube either could have become a pure user generated platform not engaging with big biz, and strictly following DMCA only : counter notice is not for Google's to review merit , claims like above are penalized, you have to take it court if there is a counter notice, no three strikes nonsense, no flawed content ID system - all of this is placate big biz.

Alternatively they could have become a curated content platform (like nebula) them wanting to do everything is why we are here.

Every other user content platform in video or otherwise is working perfectly fine with DMCA framework, they all have considerably less resources than Google.

P.S. Google's inability to put people to support content creators and this dispute process even for creators with 100' of millions of views is simply about Google culture of not believing in user/creater support (product) , big biz sure gets human support.

This is the evil of monopoly. Monopolies destroy freedom and will bring back feudalism. The GAFAM must be broken up.

Yannis Varoufakis discusses this in this (oh the irony) youtube video: https://www.youtube.com/watch?v=zlkuA1mx0pc

My daughter plays guitar and violin and this is a very common issue. Moreover Youtube automatically labels some original performances as if they use third-party soundtrack. Both annoying and flattering.

I’ve been working on a project converting my piano curriculum into a 42 week YouTube series (currently on week 34, almost there!) and one of the first challenges I ran into, other than sucking a video production, was what to do about teaching repertoire so as not to run into copyright claims.

I decided to compose my own material each week and it has turned out to be the best thing about producing the series. At first the idea of writing a new intermediate level piece each week seemed daunting but it’s actually been quite liberating. Also when I’m finished with the series I’ll be able to publish the collection as my own book of repertoire.

If you’ve been teaching for a while (like 10 years) I would highly recommend going beyond the safety net of pre published teaching repertoire and try making some yourself. Find the deeper connection to what you teach about music by encoding into, well, music! New music, that reflects your unique relationship with the craft.

If this is Google being proactive about avoiding lawsuits, how come the same issue doesn't affect Patreon?

A number of 'reaction' channels have copyrighted content (full music videos) that is clipped on Youtube to avoid strikes. The full versions of these videos are on the uploader's Patreon.

YouTube is an advertising company. Period.

The problem won’t be fixed as it’s not a problem... who cares if some piano teacher gets silenced? It matters not one iota to the revenue.

You could ban 10% of ‘content creators’ and it wouldn’t matter: people watch YouTube like they used to watch TV: it doesn’t matter what’s on or how crappy it is. You’ll keep watching. Your favorite show got cancelled? You’ll grouse and watch something else... your consumption of pharmaceutical and insurance ads will not be affected at all.

There will be no competitor emerge in this medium... just as no ‘competitor’ ever emerged in the established newspaper domain.

There may be a new medium that emerges that nukes YouTube. Can’t imagine what it would be.

YouTube's system has some massive design flaws. One of which is that the default way copyright holders get treated varies based off whether you're doing automated ContentID type claims vs whether you merely own the copyright to your own work you created.

As a result, you get better outcomes if you own a piece of content-ID'd work that you include in your videos, which is completely silly. Basically, if you copyright claim your own work automatically, if someone else comes along and tries to do the same, worst-case you split the advertising revenue among all the copyright claimants. If you just upload it normally, you get zero if someone claims it.

That is the reason these guys decided to Copyrighting all the melodies to avoid accidental infringement

https://www.youtube.com/watch?v=sJtm0MoOgiU TED talk

did it help?

I can think of one possible quick-fix solution to this: They make a library/catalog of public domain media, and whenever a claim is raised, the offending media is compared to the library of public domain.

Of course, it's far from perfect. For example - even though a piece of music is public domain, it doesn't mean that the interpretation is public domain. For example, if you cover a public domain song, you can claim copyright for that performance. This is where the automated claims system goes haywire - you actually need people (for now) to distinguish between original piece and a cover piece.

Maybe the adequate counteraction is to file DMCA claims against Google itself. Like the German newspapers did. Google News copies their headlines and intro. And fotos.

Or Australian news. Google News copies everything.

Or Google Search. Googles copies everything. Text, fotos, original work, derivative work. Nobody needs to go into museums anymore. Google has it. Nobody needs go into cinemas anymore. Google has it.

Youtube is full of copyright violations, and has no adequate support mechanism to handle claims and violations, as we saw in this example. They are policing themselves, the biggest thieve of all.

Google - you are becoming the most dump entity in the internet. Laughable level of work you are providing in recent years. Maybe consider laundry business if you can't grasp technology for people.

https://youtu.be/-ThyLB6bakk is the referenced "Wicca moonlight" performed by Alice violet molland

This is widely known among the classical musician community. Nearly every time I upload a YouTube video of me sitting at my piano playing something written 150+ years ago, I'll get a copyright strike.

In the past, I'd simply affirm it was mine and it would go away.

It looks like the new system isn't so forgiving.

This probably won't change until the right person gets a claim and hires the right lawyer who thinks of a good argument to win a massive judgement.

Well remember when they scanned a bunch of books? Like all the books in a bunch of libraries?

That was physically turning pages, photographing them, doing OCR and then correcting that. By comparison create a library of public works in terms of audio to test copyright claims, and reject or put greater requirements on ones for public domain works would be relatively easy.

It's google. I can live without it. Do you?

What about copyright it self? If they have manage to abuse it even with google what are chances that they would not find a way to abuse it on smaller platforms?

In this situation here google is not willing to check if an automatism failed. I do not think that problem applies to selfhostetd content. But who knows, its a world where everything is possible....

They would. That's why you should avoid platforms, they go against the principles of the decentralized World Wide Web.

I sometimes wish HN had a way to summon one of the apparently very few sane and effective people at companies like Google, so that they could at least feed stuff like this into the internal task list for future attention.

One can hope it happens, but it would be nice to know, and facts indicate it does not happen enough.

In this case, a solution is to match against obviously known public domain first, and failing that, check copyright claims. I’d imagine google has or can quickly build the necessary db of all public domain music, starting with music from 100+ years ago. Also identify copyright claims that conflict with that db ahead of time.

Could we fix this (partially) by writing software that generates every possible piece of music and then upload it to YouTube? If you never claim copyright on the pieces, No one else could claim them in the future cause you uploaded it first. Now we only need a way to filter out all existing copyrighted music I guess

The current copyright law is unsustainable. Pretty much every melody you can think of is copyrighted and as an artist you can be anxious that anything you make can be claimed by someone else. Copyright laws need urgent change as they are not fit for purpose...

In this case that's not the problem. Beethoven isn't copyrighted, but public.

Yeah but certain Beethoven recordings are copyrighted. Legally you are totally allowed to play his works on an instrument, while you could still break someones copyright by e.g. using their 2019 recording of the work which they sell on CD or whatnot.

How will sonic fingerprinting be able to decide whether it is that copyrighted work, or whether it is somwonw playing it live? It just won't

The problem is that Youtube is pretending that their shoddy sonic fingerprinting system works and is fine with 3rd parties using their dumb system to defraud their users

Yeah, I think they should mark recordings where the melody is public domain and review these cases manually.

It's still related to copyright law.

If the law promotes, say by making it a default, that one side can easily benefit from a certain behaviour, then people, companies will assume that as the behaviour to favour. And then these are the consequences: that going around claiming rights on whatever you feel like is not only not punished by Youtube but carelessly accepted. Because it's just easier, less effort.

I agree. On top of that you have bands from 50 years ago suing each other because two riffs sound vaguely alike, its ridiculous.

Copyright cannot be fixed, it should be abolished. It made sense in the era of physical books and huge printing presses. That era is long gone, copying have made copying trivial.

That seems backwards: the concept of copyright was introduced because the printing press made it so much easier to make a large number of exact copies. If you agree with the introduction of copyright in that era, as you say you do, how does it then make sense to abolish it again when copying becomes even easier?

Printing presses were not ubiquitous. There were relatively few of them, not everyone had access, there were limitations to the scale of copyright infringement due to the physical nature of books. People had to be major industry players in order to infringe copyright at scale and it was necessary to do it for profit in order to cover the significant costs involved.

Today virtually everyone has access to a computer that can create and transmit unlimited copies of any data at massive scales and at negligible costs and there's no way to stop it from happening unless you destroy computing freedom by making it so processors refuse to run software not signed by the government.

Copyright needs to go away because the alternative is the total destruction of free computing as we know it today. I want a future where I'm in control of my devices and can write my own software if I want without the need for some government license. If the copyright industry must die in order to protect that future, so be it.

https://www.youtube.com/watch?v=xDcvPf78g1k&t=12s&ab_channel...

The irony and hypocrisy of this are incredible.

IP is such a scam.

In ten, maybe twenty years I predict storage and cpu will be so cheap that everyone can just host their own data in their own router, and provide some universal API to it. Many problems solved, including privacy.

We've already had the capability to do this for many years. See Napster or BitTorrent.

The thorny issue is that legitimate copyright holders lose out when anyone can freely host and share their work. That leads to discussions about whether there should be copyright at all, whether digital creators have a right to charge for their work, etc. Those are the problems that need to be solved, not the technical issues.

Why don’t people upload the video from one account, and then claim copyright infringement from another account, and then intentionally drag out the appeals process as a form of defense against baseless claims?

My daughter got her first content claim when she was a 7 year old violinist. Proud father moment.

Dispute was allowed to lapse.

I’m on day 22 or so of another dispute from my son’s violin performance. I don’t see this going anywhere either.

The irony is that Beethoven probably stole it. Even to this day, music composition has always involved "something old, something new, something borrowed, something blue," to quote Duke Ellington.

How come this video is not interrupted by ads?! Something is fishy here.

I think uploaders can choose whether to include ads in their videos.

Someone needs for form a defense association which indie artists and content producers can join/pay into. Use this structure and funds to defend artists against these claims.

https://youtu.be/-ThyLB6bakk

Here is the “wicka moonlight”. Already 6.1K dislikes. I wonder why...

edit: Every. single. video I uploaded to youtube has been copyright claimed.

If you know a reasonable alternative to youtube, please let me know.

Google might be doing this to show the negative aspects of the current copyright system. Maybe they want to annoy enough people so that something might change.

This should bear consequences for YouTube as they have enabled such brutal treatment on this woman and many others.

YouTube being hostile towards their creators as usual.

Is this all because of bad American law, or is youtube proactively being bad in their moderation process?

It's a problem of scale. It's perfectly possibile to manage a few hundred disputes on a daily basis, but with >5000 uploads per minute, it becomes impossible to do manually.

The worst part is the asymmetry between claimants and creators. But that's the fucked up nature of civil law - the burden of proof is reversed.

> It's a problem of scale. It's perfectly possibile to manage a few hundred disputes on a daily basis, but with >5000 uploads per minute, it becomes impossible to do manually.

Nonsense. Assuming the revenue to Google increases linearly with uploads, so does their ability to hire content moderators.

Revenue is a function of views, not uploads.

This will get worse in the EU when the upload filters are in place.

Upload to a decentralized service like PeerTube or Odysee instead.

Reply on news.ycombinator.com