YouTube can now warn creators about copyright issues before videos are posted | Hacker News Reader

YouTube can now warn creators about copyright issues before videos are posted

theverge.com

151 pointsfinphil112 comments

"If YouTube’s copyright identification system finds a violation after a video is scanned, the rights holder’s policy will be automatically applied to the video, according to the company."

During the height of the quarantine last year I managed the live-streaming of the traditional form of the Catholic Mass (1962 Missal, substantially the same ruberics and ceremonials from the fifth century until the 1960s) at a local parish and on EVERY Sung High Mass there were multiple copyright flags applied because the liturgical hymns sung by the choir, the "newest" of which averaged 250 years in age and some of which were over a thousand years old, were part of a commercial recording by some professional choral ensemble. Had someone played a CD of that professional recording in the chapel then this would have been a copyright violation, but it's OBVIOUS to any human who listens to the sound on those videos that this was a live performance -- by volunteer, part-time singers who make lots of mistakes -- of works that have been in the public domain for centuries.

This system is clearly a CYA for Google: they would rather flag videos and PERMANENTLY demonetize content creators based on repeated false positives (but still show advertisements on the videos anyway!) than to risk even the remotest chance that a content creator might sue YouTube for enabling copyright infringement.

...even the remotest chance that a content creator might sue YouTube for enabling copyright infringement.

Does this happen often? Ever? How does one sue a firm that never has humans available for interaction with the public? One suspects this is the motivation they'd like us to credit, because they have other, less defensible, motivations.

Fwiw, you sue in public courts where Google loses by default if they don't appear. Google does not yet run its own parallel court system and star chamber.

EDIT: On second thought, if their services require binding arbitration, they do. Arbitration in contracts of adhesion should be flatly illegal.

Would a copyright owner who doesn't use YouTube be subject to such binding arbitration? Presumably such an owner could hire a third party to discover hypothetically infringing content?

All parties involved would have to have agreed in writing to binding arbitration, so yes any non-party can do whatever they want.

EDIT: IANAL and this is not financial advice. ;)

I don’t think people realize how companies/people use binding arbitration to their advantage?

These companies pay the arbitrators. It’s usually done by mail, and their usually is no investigation into your case.

The little guy wins only in egregious situations, and I have never met anyone who won especially going up against a big company?

I get arbitration agreements for small companies, but not for huge monopolies, and scummy high interest rate card it cards.

I had a run in with Columbia Credit Services, and still have a judgement following me over a credit card. Someone decided to look into the arbitration process of this company, and something like 99.9 percent of the time the customer lost. Even hired process servers whom would lie about serving the papers. All over a credit card sent in the mail.

(Judgements last 10 years, and can be renewed. They tick away at 10% per year, if the creditor can’t immediately your life miserable, and take your stuff with a warrant.)

I think it can be okay with companies large or small if the contract is something that is negotiated with real opportunity for making changes like removing such agreements. Arbitration when done with the real consent of both parties and with the choice of a neutral party that isn't expecting more business from one of them can save time and money. However, it's tyrannical for both large and small companies to just force these things on people regardless of how vulnerable their business model is.

I'd note that it has never occurred to me that arbitration would ever have been in my own interest.

Are you asking if a Record Company would sue someone on the Internet for copyright infringement? They are rather famous for doing so, and more than that, for asking for extremely large settlements. Sometimes calculating the "damages" in completely ridiculous ways like multiplying every view by $50,000.

Don't forget what happened to cdrom.com. Literally sued out of existence because they allowed people to upload their own music to the cloud. They couldn't even share it, it was just uploaded to the cloud. Business completely destroyed.

The way cdrom.com worked is you would upload a fingerprint of your music. Then if they already had a track that matched the fingerprint, they would link that to your account and not require you to upload the whole track. So this was technically not within the scope of what is allowed by copyright law, which is why they got in trouble.

Which is kind of strange how when something is "technically" a violation, even if it is within the spirit of the law (or equiv. to what is allowed), the copyright hammer comes down. Whereas when something is "technically" NOT a copyright violation, the hammer still comes down (such as that company that used a bunch of tiny TV antennas to stream broadcast TV over the internet to other markets -- each customer was "renting" one of the dedicated TV antennas and using the internet as a "long cable". Or the company that would dedicate a DVD player to each customer and stream movies across the internet, again claiming the internet was just a "very long cable". They both got shut down hard even though they weren't "technically" in violation).

The whole "they're not physically your bits" argument never made a lick of sense. They never would be after you uploaded them over the Internet.

The whole case depended on the Doctrine of First Sale not applying to CDs, so that's what the courts agreed to. The record companies could have gone after second hand record stores, at least the ones that sold CDs, if they wanted to at that point but it was kind of unnecessary because they were already dying.

The reason why cdrom.com was sued out of existence was that it became trivial to simply publish directories of fingerprints and upload massive numbers of fingerprints into your account and get access to music you never bought.

So if they had an algorithm that would ask for a unique fingerprint (say by sampling random parts of the track), then they would have been in the clear? Or, more accurately, would the record labels have been more likely to strike a deal if that was the case?

The record companies wanted them dead. There was no negotiating. Remember this was back before iTunes. They saw streaming as only a threat.

In USA civil courts the wealthier party typically wins. Alphabet is going to be the wealthier party in most cases. Certainly more than "cdrom.com". This was one of the advantages cited back when Google purchased YouTube.

https://archive.thinkprogress.org/poor-people-dont-stand-a-c...

So why does Google bother with the false-positive heavy with draconian penalties ContentID stuff then?

Oh, because Record Companies are still bigger than people who create content for YouTube.

Yes, "why" is a great question to ask, because ISTM Alphabet benefits from being able to blame an essentially unappealable process while it curates content uploaded to its platform by creators. I'm a rancher, and like Alphabet I prefer not to get in extended arguments with my livestock.

If blame for the excesses of curation couldn't be offloaded to third parties, creators could form a class for enterprising attorneys.

Sure, it's also good to have a ready defense against litigious record companies, but large firms don't have to have just one reason for doing things in particular ways. It is easy to imagine an alternative to the "ContentID" system that would be less capricious and would still make record companies happy.

I think it mostly boils down to business loss that they would otherwise incur if they set themselves against those companies and thought them tooth and nail (the actual lawsuit cost would be insignificant). Meaning, no Google/Youtube Music deals with Viacom and potentially other RIAA members.

I think you're thinking of mp3.com. Cdrom.com is the Walnut Creek FTP site.

Viacom International Inc. v. YouTube, Inc. is basically what caused Content ID to come into existence, despite being ruled in Youtube's favor. It was going to go to appeal but a non-monetary settlement was agreed upon, which is why it's believed that Content ID came from that.

[1]: https://en.wikipedia.org/wiki/Viacom_International_Inc._v._Y....

The copyright owners are terrified by a new medium they don't understand for the most part so they extract their rents the way you know how.

I made a dumb mashup containing spoken words and a song, I'm not sure whether that is fair use or not, but it got copyright striked for infringing upon something like "test song #3" by some record label. I couldn't even find out what that was.

> How does one sue a firm that never has humans available for interaction with the public?

You don't sue a company by contacting customer service. I'm pretty sure they need to have some official representative on file somewhere, and you could probably serve them the lawsuit by sending a certified letter to their official address.

The bigger barrier is Google/Youtube almost certainly have an arbitration clause in their terms of service.

Music copyrights are a minefield, even for old music, because it is often performed and rearranged, both of which would restart the copyright clock. And even if your church pays for performance licenses, it may not apply to when Google redistributes that performance. There is not a simple solution to the problem, even if there are 'mistakes' in the performance.

Mistakes only make it more impossible of a problem to solve, because 2 notes could be the difference between a public domain work, and a rearrangement done last week. Did the performers make a mistake on the two notes that differentiate the works, or a different place? It's simply not possible to tell from the audio.

"Music copyrights are a minefield, even for old music, because it is often performed and rearranged, both of which would restart the copyright clock."

Most of what the choir was flagged on was Gregorian Chant, the settings of which trace their origins to Old Testament worship in the Temple of Jerusalem, possibly as far back as King David's tabernacle singers. The texts are all ancient (200 to 1600 years old) and the "new" settings were composed by William Byrd (1540 - 1623) and Palestrina (1525 - 1594). We're not talking about something that was recently reset... though that would explain why Morten Lauridsen's "O Magnum Mysterium" is almost never heard in Divine Worship (that setting isn't even 30 years old yet).

There are undoubtably a ton of variations of those chants over the past 120 years and probably millions of recorded performances of those chants. Any of which would be subject to a new copyright on the date of that performance or rearrangement, regardless of the original date of the composition.

A work being widely performed is a common situation that complicates attempts to solve the problem automatically. An infinite number of monkeys randomly singing a song will eventually sing one that sounds exactly like yours. How does an algorithm determine the difference in intent between my randomly singing choir of monkeys and your church's performance? It can't, as it compares audio -- it doesn't reason like humans do. Content ID doesn't know who your performers are, who recorded it, which sheet music your performers were reading from, where they made mistakes, and/or what licenses you have. (and FWIW, any given human watching the video wouldn't know much of this either)

Two audibly identical works can be non-infringing, and two audibly differentiable works can be infringing. ContentID is an approximation -- it does not and never will have enough information to accurately judge everything.

That argument doesn't hold water when my Pixel can hear a piece of classical music and identify the particular performer/album, not just the title of the music itself.

If my phone can do that on its own (Google claims they aren't sending any audio to the cloud), then Content ID can surely do just as well, if not better, and not flag someone's church choir's performance as some recorded, copyrighted work.

You are missing a big part of the problem space as it pertains to copyright law. That application is fingerprinting a portion of the audio and looking it up for a match in an existing database of copyrighted works for a match. This is the same thing that Content ID does, and it's exactly why it has false positives.

Knowing whether portions of audio in a recording match other existing recordings is not enough to determine whether or not something breaks copyright law.

You can demonstrate this with Shazam (/whatever the Google one is called). Live performances of any decent quality will often match recording fingerprints in their DB and give you a result of a particular artist/album.

The law is simply more complicated than `if(audioMatches) then {illegal}`

Sometimes a match on audio is not an illegal copyright infringement.

Sometimes audio that doesn't match is illegal copyright infringement.

Let's say you play a song from the year 1200 on the violin. You have the copyright to that performance. Now, let's say I play the same song, and I play it relatively well. I have the copyright to my performance. Likely, the audio fingerprints match, and Shazam/Google/Content ID will all say that my recording matches yours -- even though my performance was entirely legal, and we both own copyrights to our respective performances.

The information Google is missing is:

1. it doesn't know that the sheet music we were both reading was authored in 1200. It doesn't have this information, it just has the audio.

2. it doesn't know that a different human physically performed the work. It just knows that the audio is the same.

I wonder how long those portions of audio are.

With portion length and bitrate there is some theoretical number of music samples that could be 'brute forced' to cover the entire content ID space.

I wonder if their snippets are long enough to make this an impossible number, or if this theoretical attack on Content ID could be put into practice.

>The law is simply more complicated than `if(audioMatches) then {illegal}`

>Sometimes a match on audio is not an illegal copyright infringement.

>Sometimes audio that doesn't match is illegal copyright infringement.

>Let's say you play a song from the year 1200 on the violin. You have the copyright to that performance. Now, let's say I play the same song, and I play it relatively well. I have the copyright to my performance. Likely, the audio fingerprints match, and Shazam/Google/Content ID will all say that my recording matches yours -- even though my performance was entirely legal, and we both own copyrights to our respective performances.

>The information Google is missing is that it doesn't know that the sheet music we were both reading was authored in 1200.

I'll be honest I don't know anything about copyright but wouldn't solution be for example you as a YouTube content creator can claim the right to that particular song and the other guy as well and know Google has a record in their database that says:

John owns right to 1200 violin song (song id)

Jack owns right to 1200 violin song (song id)

And then IF John's (song id) matches Jack's (song id) don't send copyright strike.

Google needs to build YouTube knowledge graph so it has better awareness on information.

The problem is, Google doesn't know they both have the rights to it. It just knows they both uploaded it. They might neither of them have the rights to it. It's just some arbitrary audio, and google was able to find another very similar audio somewhere else.

As I understand, that already exists today, and you can claim that you own a piece of content through the content ID system, by disputing the match.

The problem is that most people uploading content do not have legal council helping them with this, so they:

1. don’t know that they can/should dispute

2. are afraid of escalating the situation to court

3. ignore it because legalese is scary/uninteresting and they don’t care about filling out forms over it and just want to upload their dank memes

Is it legal to put consumer / creator in that position vs goliath? It seems that law needs changing. If a creator wants to dispute the claim, Google should pay for their independent lawyers if they can't afford it.

I've watched some YouTube videos that show the process of disputing a Content ID match. It's a pretty straightforward process and Google explains how it works.

https://support.google.com/youtube/answer/2797454?hl=en

But Google can't fix legal illiteracy. They warn:

> Before you dispute: Make sure you understand how fair use and the public domain work. YouTube can’t help you decide whether you should dispute a claim. If you’re not sure what to do, you may want to seek your own legal advice.

I don't think it would really even help if lawyers were provided. All of the information needed to understand the situation is widely available. But most people can't be bothered to RTFM. They just want to upload their stuff any click 'next' without reading anything.

Can't speak for YouTube, but we[0] solved issues like this by making public domain the first class citizen of the system. Once a work enters a public domain in any of the 195 countries, is no longer possible to register it in our system for that country. This prevents any rightsholder to abuse the system.

It also allows platforms and creators to freely utilize public domain, royalty free and Creative Commons works in their content without worrying about situation like this. We also distinguish explicitly between publishing and sound recording/performance so one can register one without impacting the other.

[0] https://pex.com

This sounds like a sufficiently hard problem to devote a whole business to.

If this is not a sarcasm then that’s what we did. If it is, I honestly believe that this will end up being better that it seems on the surface.

We modeled the company on VISA. The idea is to bring three-sided marketplace to allow creators to do what they do best, create, without worrying about the laws and regulations across the world, platforms to bring new opportunities to express to creators and rightsholders to benefit from their content.

Not sarcasm, though in retrospect I can certainly see how my choice of sentence structure would indicate that. Seems like there's a lot of value here.

Thank you. Appreciate it

Do you have any public pricing or startup plans?

The way we make money is we charge % of revenue generated from the licensed content. So starting is free, using the service is free and the only time one would pay us if the platform makes money from any licensed content.

So if I am launching a Spotify/Audible-style music/audiobook streaming platform, how will the pricing work out? Do we pay you instead of the original author for any user uploaded content? If the original author chooses to upload their content onto our platform, do we scan it with your API and explicitly whitelist it and pay them directly?

Your company looks very cool btw. What's your email?

My email is r@pex.com

The way this works is that you send all content through us via our SDK which generates fingerprints (so content doesn’t exchange hands). If the rightsholder is registered with us (it’s free), then we can take care of everything from payments, ADR (alternate dispute resolution), reporting, ...

If the rightsholder uploads their own content to the platform, they are able to pre-license their content in our system (different word for allowlist, but a bit more nuanced as these lists run into GDPR issues).

Rightsholders are also free to have direct relationships with platforms while we run the backend. We don’t require any particular licensing to be done through us.

Hopefully this makes sense.

I might suggest using a different email for public communications.

When I read your email, it looks like "rape x dot com", which is quite unfortunate.

I must say that’s the first time someone said something like this. Not to dismiss your comment, but I don’t think this is a common thing.

I think one could also read it as rat pex, but also not something that ever happened before :)

I, uh, also had the reaction as GP…

Visually the first 4 letters do look like rape, r@pe x.com. Rat x would be uncommon since there is no ‘t’ to be seen there, and phoentically its a long R, sounds like “are at pecks dot com”. From a marketing perspective I’d consider changing it just because of the negative connotations the word r@pe elicits in the mind. But if it’s personal email address then not much reason to consider my opinion at all.

P.s., if it is a personal email, please do disregard my message, and feel free to in return critique my own username: vgb2k18, it’s an acronym for very good boy 2018.

Fwiw, "rape x" is exactly what I read your email as before I read the other comment saying the same.

Dude, change it. It’s obviously an unintentional thing, but it is a thing.

If penisland.com (formerly Pen Island) and therapist.com periodically appears on unintentionally funny URL lists regularly, you might find your email address getting raised eyebrows.

I thought the exact same thing fwiw.

> This system is clearly a CYA for Google:

Sorry, the DMCA protections for google are almost absolute. Without a notice and takedown they are unexposed.

Their handling might just be incompetence, or it might be a money maker (e.g. to force people to allow copyright holders to monetize their videos)... what it isn't is a CYA.

YouTube's copyright system isn't the DMCA.

The DMCA contains minimum protections for abuse (penalty of perjury), whereas YouTube's private system contains absolutely none. Content owners literally make money by flagging as much as possible, and there's no reason not to.

No, that's not correct. Google goes above and beyond what the DMCA requires in order to curry favor with the movie/TV/music studios. Presumably Content ID was a part of Google's bend-over-backwards settlement with Viacom when they sued.

The DMCA process does not require content hosters to pro-actively guess whether or not newly-posted material might violate someone's copyright. It just requires that they provide an avenue for rightsholders to file claims, which have to be promptly acted upon.

Google's system unnecessarily favors big players at the expense of individual creators. For all the flak the DMCA gets, its claim/counterclaim process is way better for creators than Google's system is.

>For all the flak the DMCA gets, its claim/counterclaim process is way better for creators than Google's system is.

I suspect this is the crack in the YT foundation that will allow for real competition to emerge. Several YouTube creators have formed a competing network specifically because of the BS they deal with regarding ContentID being so anti-creator.

What is the competing network? Interested in YT alternatives. Can’t wait to see their medieval copyright process go.

Probably nebula: https://watchnebula.com/

Discovery is really horrible though, at least in my opinion.

You get a free subscription if you sign up for curiositystream: https://help.curiositystream.com/hc/en-us/articles/360034158...

I got the promotion from one of the legal Eagle videos: https://youtu.be/BnjhHPU-uRw

I want to use Nebula, but the UX there is such garbage (less reliable streams, no discovery/algo, no good movile/tx ux) it's really hard to get onboard, even if I'd prefer to.

It's a bit more complicated than that.

Google's system is optional for both the content creator and the rightsholder. If the rightsholder would prefer to use the DMCA, they obviously can. If the content creator would prefer the rightsholder use the DMCA, they can dispute+appeal which will force the rightsholder to file a DMCA takedown if they want to continue with their claim.

Unfortunately there's often a lot of confusion about the dispute+appeals process, since it's a multi-step process and there's a ton of people incorrectly conflating disputes with appeals, but the important takeaway is that a content creator can choose to force a DMCA takedown via an appeal (which happens after a rejected dispute).

The reason content creators rarely choose to go the route of forcing a DMCA takedown is because:

1) the only remedy for a DMCA takedown is a removal of the entire video, which is often less preferable than the remedies offered by the google process (ex: replacing/muting the audio in the claimed section, monetization split, restrictions in only some geographical regions)

2) the DMCA repeat infringers clause requires that companies terminate relationships with people that repeatedly get DMCA takedowns. This manifests on YouTube as a 3-strikes-and-your-channel-is-deleted policy that only applies to DMCA takedowns, but not copyright claims that come through the optional google process.

3) There's some scary warnings about how particularly vindictive rightsholders react to this by choosing to file a lawsuit against the creator in addition to filing the DMCA takedown

> Google's system is optional for both the content creator and the rightsholder.

So, optional in the "or we'll break your legs" sense.

I suppose. It's optional in the sense that the DMCA process is the alternative, and the DMCA sucks really bad.

My interpretation of google's copyright system is that it tries to straddle the line of being both significantly less punishing to content creators while also giving enough of a remedy to rightsholders so they don't end up preferring the DMCA takedown.

The system is completely useless if rightsholders disfavor it and use the DMCA instead (which by law they can do at any time). This is a really unfortunate challenge around the design of copyright systems like googles that isn't often appreciated.

All this would be so much easier if the DMCA was just reformed. There are so many obvious changes that could be made to make it much more tolerable and remove the need for complex systems like google's.

Just want to leave something here:

>the DMCA repeat infringers clause requires that companies terminate relationships with people that repeatedly get DMCA takedowns.

AFAIK there is no clause that protects the reverse process to this: copyright holders can just bulk file bullshit claims without consequences. As usual, it's the big corps that are favored by such laws at the expense of the individual.

Also

> monetization split

exists only for ad-revenue, which is split between YouTube and creators. That doesn't apply to copyright claimed material, where all money goes to the copyright holder, no question asked whatsoever.

> No, that's not correct. Google goes above and beyond what the DMCA requires in order to curry favor with the movie/TV/music studios. Presumably Content ID was a part of Google's bend-over-backwards settlement with Viacom when they sued.

I imagine it was because of this: https://en.wikipedia.org/wiki/Viacom_International_Inc._v._Y....

Yes, Google won in the end, but there were times when it didn't (they won on appeal) and the cost of the lawsuit itself, the insecurity of its outcome and the potential impact that could have down the line to doing business with Viacom I think were enough to act.

> No, that's not correct.

Exactly what are you saying isn't correct?

Are you claiming that if google does not proactively take down content, absent any DMCA complaint, they are libel for infringement and do not benefit from S512 limitations on liability?

If you are not, how is my statement that this isn't a case of google covering their ass to avoid legal liability (because none exists)?

Great to see another TLMer out on the internet :) Sounds like you have an awesome choir (schola?)!

Thanks for sharing.

If you know you're in the clear in terms of copyright, you can always dispute to clear the claim.

You want to pay attention to what kind of claim it is - it can be against the performance (recording), or it can be against the composition. If it's performance, and your recording is a live recording of obviously different performance, it's a false match and you can dispute. If it's against the composition, if you are absolutely sure you're using the original composition and not any modern rearrangement, you can dispute.

I've never lost my dispute so far.

Why exactly do people keep thinking Google cares about the profit of its apps at all? They are notorious for killing off a ton of successful products that have a ton of users. I’m pretty sure they are using selling tech products as a front for whatever they are really doing. Probably building a temple to worship that AI demon they are creating.

Meanwhile, in the real world, Jim Sterling is facing their channel being taken down because a random game developer didn't like their review and decided to copyright flag a bunch of different videos. They're at 2 strikes now, which means one more and the channel goes down. Fantastic system.

https://twitter.com/JimSterling/status/1372585399667806211

Huh... This actually seems like not a bad way to ensure that Google is aware of the problem. A sufficiently large reddit community could probably knock a lot of big fish off of youtube with this approach

The fish big enough for google to care about are all protected from this stuff.

YouTube is completely and utterly broken for anyone smaller than “I make tens/hundreds of thousands for google and have a management team to handle my YouTube issues”

Sterling does not monetize their videos, instead relying on Patreon support, so even fewer shits will be given.

This has been happening for years. A friend of mine lost their channel twice because, somehow, his channel got hit by a ban wave of channels that were uploading specific Japanese adult videos with a very specific title.

It took him weeks of messaging YouTube to even get to a human being, and even then it was another month for his channel to come back. They know the system is bad, they just don't care.

I've always despised how YouTube removes videos with no trace of what was once there. I have dozens of videos saved to playlists that I will never be able to recover because there is no identifying information. Why not take down the file but leave the title for reference?

This is why I youtube-dl everything

I YouTube-dl anything/everything I think I might like to watch again. Tutorials, reference videos, music videos, you name it. Storage is cheap and YouTube has proven they’re not a place for long term storage.

Might be a good opportunity for someone to create a browser plugin to just download all Youtube videos that they watch.

Doesn't always work, but try using the Wayback Machine for the titles.

Did you see how Twitch handles copyright issues? They're latest solution is to make it easier for streamers to mass delete all their saved streams ("VODs") to not get DMCA strikes. Compared to Twitch Youtube seem much more competent and professional.

https://www.esportstalk.com/news/after-a-year-twitchs-dmca-u...

Every company looks more competent and professional compared to Twitch as of late. The latter seems to be involved in at least one controversy a month.

What's weird about the Twitch situation is that they already had a system to solve this. If you had a VOD that flagged their Content ID, the offending part would get muted. Why not do the same thing for manual takedown requests?

It's baffling that this feature wasn't already available to content creators.

Agreed... and it makes you wonder why Google held back on this.

It would have been baffling to me in the past, but I've felt it's pretty clear that YouTube is more interested in catering to mainstream entertainment giants rather than individual content creators. It's baffling they'd even bother with this now. The only reason I can think of why they would introduce this feature is that their algorithm has been backfiring on content creators they are politically aligned with.

I’m pretty sure it was in a backhanded way. I would upload a video and keep it private. Then you could wait and see if any part got flagged. If it didn’t then you could publish.

What if they didn't want to give users a way to iterate quickly through various attempts at obfuscating copyright content by just the right amount in order to bypass the copyright screening?

Not saying that was the reason but it's not implausible.

I could think that they also wanted to increase engagement on the site. People better respond to music they know, so they were milking sweet ad commissions as long as they could. Maybe labels realised YT is taking a mickey and had a strong conversation with the execs.

I am guessing, because of the complexity of the problem. With the previous way of doing things I am guessing youtube either push all the videos that get uploaded in to some queue to detect copyright violations, or once a video reaches a certain number of views the copyright algorithm gets triggered.

With the new system I am guessing they have to do this thing in real time and almost anything real-time is a complex problem to deal with right?

This should have been the norm & if someone else copyrights an older content then the new video is to be scrutinised not the creators content & otherwise a notice needs to be shown instead of a strike outright for valid claims

Big labels are churning music at an unprecedented rate as if they wanted to copyright all possible combinations of chords / melodies (that universally sound good). At one point in the future it won't be possible to upload any new music without having to pay those label or just being outright censored. This is going to be a disaster for culture in few years and winners will be big corporations.

> This is going to be a disaster for culture in few years and winners will be big corporations.

Yes, that's the point. Can you even comprehend how much money two corporations will make off governments and individuals?

A problem about copyright I encountered:

I don't hear and need captions for videos. Sometimes material I have is not captioned. But when I resubmit them I can get a copyright warning. I stopped resubmitting to avoid Youtube blocking me. I could create throwaway accounts, but this is a hassle, and it's only me who benefits from captions.

This is a case were indiscriminate enforcing of copyright is hurting people with disabilities.

Honestly this was too obvious to not do anything until now. Next thing is fixing the demonetisation bs

The only reason why they done is because it's mandatory under the impending legislation in EU [0].

[0] https://en.wikipedia.org/wiki/Directive_on_Copyright_in_the_...

So it is the might filter and pre-cursor of preventive censorship in the EU. Marvellous.

Not sure what are you trying to say, but no law gets anything right. That said, the current market is very one-sided where platforms abuse both sides (creators and rightsholders) while nobody can do nothing about it. So I welcome the governments trying to find balance, even if it’s not perfect.

I think the law is cutting way too far in one direction, but I believe this is the first time the market has an opportunity to work together to find some form of balance.

Everyone has different standards. To me this is unacceptable - we used to pay so much money to all those unelected people and their duty was to get things right for citizens, not for big corporations (who likely also pay them through unofficial channels). I am glad that my country decided to not implement that in full once we have left the bloc.

I’m very biased here, but I think the law is going to eventually be net positive not because is so good, but because there is another (DMCA in US and E-commerce directive in EU) that allow platforms to use content without compensation.

This is unsubstantiated in increasingly connected world where digital content (and rights) are structural to its operations and business. This law is far from perfect, but it’s seeking to remedy this imbalance. I agree that this could’ve done better, but we live in imperfect world with lobbing being part of the process.

So my hope and believe, one I built my company around, is that this law will allow to create balanced marketplace (similar to VISA) that allows all 3 sides (creators, platforms and rightsholders) to participate in the upside rather than focus on a zero-sum outcomes as it’s today.

I might be naive and completely wrong, but this is what I’m personally working towards.

Will YouTube also warn creators when Telemundo or extortion trolls try to claim original works as their own?

What if the video is using copyrighted material on fair use grounds? I guess you're still screwed then?

They were forced to this because of the impending legislation in EU [0] which forces them to do so. Germany is toying with what they call pre-filtering, which would allow creator/user to state fair use/exception prior identification. Essentially one could claim fair use before a rightsholder can make their claim.

So far it's part of the draft but we don't know if it will make it to the final legislation.

[0] https://en.wikipedia.org/wiki/Directive_on_Copyright_in_the_...

A hell of a lot more cooperative approach to help everyone avoid unintentional messes is long overdue.

The folks who get their content taken down I'm sure would be happy to cooperate most of the time ...

Really sad that it comes this late / after this much hassle.

This should've been implemented earlier it would save a lot of users time and hassle. Google has enormous engineering manpower and computing resources they can do pretty much anything.

I believe that artists should be reimbursed, but how YouTube does it does not look just at all. It is a simple money grab by fat major label execs who were able to cut deals with YT at the expense of regular people. Yes, if someone uses a copyrighted song, they should either be asked to use something else, provide proof of a deal or agree to send a split to an artist. I understand currently it is that the label takes all. I wonder why there isn't any lawsuit around this?

> I wonder why there isn't any lawsuit around this?

There's been a ton of action around this, including an active class action lawsuit. Congress was hearing testimony on this exact subject back last year and there's legislative action going on in the EU. I suspect this move was in direct response to those hearings.

This is a huge improvement because very often by the time ContentID claims are resolved most of the subscribers and other early-serge viewers have seen the video so that money has basically been stolen from the creator and that video will not bring in much profit after that initial surge.

It obviously isn't perfect but at least this lets the creator appeal/update the video before effectively cutting a large check to companies who very often don't deserve it.

Sounds almost like Sheckley's Watchbird already :)

Are there any copyright on watchers comments?

I can't imagine the incredible amount of man-hours that were wasted on detecting if your video has fucking copyrighted material in it. Absolute shame.

The technology has existed for a nearly 20 years.

Shazam [0] existed long before smartphones and could identify songs based on 10 second samples. YouTube can easily leverage this song-identifying technology to detect copyrighted material.

[0] https://en.wikipedia.org/wiki/Shazam_(application)

Recognising a song is not the same as recognising copyright.

Of course not, but if it can recognize the song, and the record labels have given YouTube a list of copyrighted songs, it's not too hard to put the two together.

If you accept the false positives that impact lots of us, then sure, it's not too hard. This is one of those cases where 99% is easy but society needs 100% and that's not as easy.

I'd like to see a total energy comparison between Content ID and Bitcoin PoW.

Not just them. There are companies that exist primarily to help content creators manage this stuff. It's insane.

Off-topic: this is one of the few comments you have that aren't dead.

Reply on news.ycombinator.com