YouTube's $100 Million Upload Filter Failures Demonstrate What A Disaster Article 13 Will Be For The Internet

from the your-future-internet dept

The entire Article 13 debate is a weird one. It appears that both the recording industry and the film industry are going for broke on this one. The lobbying on this started a few years back, with the rather clever but completely bogus idea of the “value gap.”

In case you haven’t followed it, the idea of the “value gap” is that (1) YouTube pays less to musicians record labels than Spotify and Apple Music do for streams. (2) YouTube’s general purpose video hosting platform is protected by intermediary protection laws (DMCA 512 in the US, Article 14 of the E-Commerce Directive in the EU) allowing users to upload whatever they want, and YouTube only has to takedown infringing works upon notice. (3) Services like Spotify and Apple Music license all their works. (4) The “lower rates” that YouTube pays must be the result of the safe harbor, and the difference in payments is the “value gap.” Article 13, then, is supposed to “fix” the value gap by completely removing any notice-and-takedown safe harbor for copyright-covered works.

Of course, almost all of this is bullshit. YouTube is used in very, very different ways from Spotify and Apple Music. While YouTube does have a competing music streaming service that is similar to Spotify/Apple Music, its payment rates there are equivalent. But on the general open platform, the rates are different. This is not because of the safe harbors, but because people use the platforms very, very differently. People use Spotify/Apple Music almost like radio — to put on music that is constantly streaming playlists of songs. YouTube has all sorts of content, most of it not music, and while some may use it as a radio-style experience, that is fairly rare. And the recording industry has always received different rates based on different platforms and different kinds of usage.

Meanwhile, Article 13 will do nothing to solve the “problem” that all the “value gap” people keep insisting is a problem. That’s because Article 13 will basically require an upload filter that will spot infringing works and block them before they get on the site (there’s more to it than that, but that’s a basic approximation of what the law will require in practice). Basically the only company that has actually done this already… is YouTube! YouTube has its ContentID system, which it has spent over $100 million developing, and which can block uploads and pull down content.

And… let’s take a look at just how much damage such a system causes. Remember, YouTube has spent more on its filter than anyone else (by far) and it is considered easily the most sophisticated and advanced such filter.

And it sucks.

Last week, I saw musician Dan Bull (who wrote/performed the Techdirt podcast theme song) complaining that he had he had received a copyright claim on a video that was his own work, and from someone whose work was not in the video at all:

The issue, apparently, was that both Dan Bull and the claimant — another independent rap artist — had used used the same instrumental, that was available on a “non-exclusive” basis, meaning lots of artists could use it. That indie artist then tried to monetize their own work, but ContentID found anyone using the same properly licensed instrumental and gave people a strike.

And, even the way in which YouTube communicated with Dan Bull was ridiculous:

If you can’t see that, YouTube told him that “your copyright dispute is currently being reviewed by the copyright owner,” to which Dan Bull rightly pointed out that’s bullshit, because it’s not “the copyright holder” who is reviewing it.

A day later, someone else pointed me to a nearly identical situation with another super popular YouTube musician, The Fat Rat. A third party claimed copyright on his song and YouTube rejected The Fat Rat’s dispute.

And, then, soon after that, he posts a similar frustration to Dan Bull when YouTube incorrectly tells him that this is in the hands of “the content owner,” when that’s not whose hands its in at all:

That’s just two examples of YouTube’s $100 million upload filter system claiming copyright on music incorrectly that I randomly came across in just the past few days. But it’s not hard to find tons of other examples of this happening. Here’s someone pointing out that each and every one of his videos is being copyright claimed one after another:

And that seems to happen quite often too. Here’s an incredible video from a popular YouTuber explaining in great detail how every single one of his videos had been copyright claimed:

In case you can’t watch that, basically the guy legally licensed some music to be used as his outro music (paid the musician $50 for it, as per their request, even though the musician later claims anyone is free to use the music, so long as they include a Spotify link to his stuff). But, then a bunch of companies get involved, including Distrokid and Audiam, who claimed basically every single one of his videos, and his email conversation with both companies is maddening, as they don’t seem to understand or care about what’s going on — and basically tell him that every time this happens he would need to email them again and hope they remove the claims, but they refuse to whitelist his account.

And here’s another person highlighting that all 3,000+ of his videos just got a ContentID copyright claim:

And here’s someone who got a copyright claim on a private, unlisted video that the uploader needed for a class assignment:

Those are just the first few results that pop up for a quick search on Twitter of “Youtube” and “copyright.” There are lots more. Indeed, there’s an interesting one from the super popular YouTuber @Jack_Septic_Eye (Sean William McLoughlin) who explains that he accidentally caused a ton of bogus copyright claims by switching YouTube networks:

In short, the $100 million ContentID system is a total fucking mess.

Bring that back around to Article 13, which will create massive fines should YouTube fail to catch any copyright-covered works with (per the industry’s wishes) no safe harbors for platforms, and this problem will only get significantly worse. Under Article 13, YouTube would be crazy to give anyone even the benefit of the doubt if they receive a copyright claim. Not only that, it would be effectively required to not just block all of these works, but to prevent many of them from ever being uploaded in the first place.

What the various people pushing for Article 13 don’t seem to realize is that bogus copyright claims happen all the time. Or there are mixups around content that multiple people license. Or there are multiple middle men making claims on behalf of copyright holders with no knowledge of who has licensed and who has not. Article 13 doesn’t take any of that into account. It just says “license everything.” Of course, the big labels/studios don’t give a fuck if this leads to censorship of you piddly little independents and amateurs. They just want YouTube to hand them a giant check to “license” their entire catalog. That’s the real endgame here. That it would block out independents and amateurs (i.e., competitors) is just icing on the cake. And, of course, it ignores that if YouTube spent $100 million on this system and it already has so many problems, just imagine how poorly everyone else’s mandatory filters will work.

Article 13 is a horrible solution to a “problem” that doesn’t even exist. The fact that YouTube’s $100 million ContentID system is so full of bogus claims shows just one of the many problems with filters — and making them mandatory won’t suddenly make YouTube pay. The whole thing is just designed to be leverage. The labels and studios want to use Article 13 as a weapon against YouTube, basically saying “give us all your money in a giant license or we’ll sue you over and over and over again.” And, of course, should they actually get that, the money is unlikely to make its way back to any of the actual artists.

Europe is doing a big thing badly, and Article 13 is an unmitigated disaster.

Filed Under: , , , , ,
Companies: youtube

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “YouTube's $100 Million Upload Filter Failures Demonstrate What A Disaster Article 13 Will Be For The Internet”

Subscribe: RSS Leave a comment
64 Comments
Anonymous Coward says:

Re: Re:

LOL Masnick’s last ditch lie is now to say that musicians don’t see any of the money so YouTube should just to be able to continue to skim and keep all the money for themselves. Except Masnick’s bullshit fails to account for the thousands of independent musicians getting ripped off by YouTube. Oh, whoops.

Anonymous Coward says:

Re: Re: Re:2 Re:

LOL you must be totally clueless about how most labels work. It doesn’t mean jack if you have a contract: the label will still do as it pleases, even with a signed contract with precisely stated royalty rate. They will just “forget” to pay the artists, even when artists “remind” them many times, ignoring them and ripping them off for eternity when they sign that damn contract, and since most contracts are exclusive it means artists cannot even take their music somewhere else.

Labels even do damage control on posts on the Internet which explain this outrageous situation, actively censoring all those which talk in a negative way about them, when it is just the truth of those frustrated signed artists that want to warn others not to make the same mistake as they did. These labels perpetrate an outright scam against artists and artists cannot even complain.

And the icing on the cake is that in most cases legal action against them costs more than the royalties which artists would have received anyway, or is not practical at all (for example when the label resides in another country) so it becomes a moot point.

Mike Masnick (profile) says:

Re: Re: Re:

LOL Masnick’s last ditch lie is now to say that musicians don’t see any of the money so YouTube should just to be able to continue to skim and keep all the money for themselves.

Hmm. Can you point out exactly where I said this… because… oh yeah, I never said anything of the sort.

It’s funny that the only way you can call me a "liar" is to flat out make up shit you claim I said that I never said. Desperate much?

Anonymous Coward says:

Re: Re: Re:

fails to account for the thousands of independent musicians getting ripped off by YouTube

Explain how voluntary use of a distribution service is being ripped off. How many of those musicians also get money from also being on other platforms, or via Patreon or other sources of patronage?

Also note, even for many musicians signed to labels, recordings are advertising for live performances, which is where they make their money.

Also note, some income is better than none, and may add up to a decent hourly rate for the actual work done in making a recording.

Cassie Fenetre says:

Re: Re:

I am continually surprised by the restraint showed by YouTube and Google over all of the legislative action being taken by European countries.

But according to Masnick (when suits his slant), Google will actually come out ahead if Article 13 is passed! My bet is that’s true and that Google is secretly promoting this. You should recall that Youtube flipped on it, without informing Masnick…

For your notion of "restraint": A) corporations are still (just barely) vulnerable / accountable to legislatures; B) IF want to be in The People’s Marketplaces and gain money, then corporations must obey rules.

I personally would have just gone straight for the nuclear option and cut of Europe from services and let their citizens raise a ruckus with their politicians.

And that’s why you aren’t in position to make those decisions. You are an immature hot-head. Your notion of pulling out of Europe is just… pitiable, really. And my bet is that most citizens of Europe would be glad if Google left. They could struggle along and develop local alternatives like Russia, China, and India, keep all money and control local, as should be.

James Burkhardt (profile) says:

This is the only reasonable decision, I mean sure Dan Bull licenced the track he samples as a backing track for this very purpose like OOTB says is all he needs to do, and the rest of the music is his own creation, but without an exclusive licence filed with the lizard overlords it is pure unadulterated THEFT for him to make works including properly licenced content! /s

More seriously, I wonder who the liability on THE Fat Rat’s content lies. Clearly, the other content creator holds liability, but I wonder if YouTube holds direct liability for proactively assisting (Content ID) in copyright fraud and appropriation (Creating an environment where the first to upload can claim the exclusive copyright on works licenced non-exclusively

Cassie Fenetre says:

So every system must automatically bar false inputs?

In what other area do you insist on that?

bogus copyright claims happen all the time.

That’s known, kid. Now YOU state what percentage of claims are bogus as compared to the daily millions of instance in which copyright works just fine.

YouTube’s $100 million ContentID system is so full of bogus claims

Exactly NONE of what you example is Youtube’s fault nor show its ID system fail!

There is NO connection here other than YOU put false claims near anti-Article-13 text! Feeble and desperate, kid.

To FIX this, Youtube needs only to follow the DMCA and restore the content when contested. Then it’s up to the claimant to continue.

Problem solved. Without any relation whatsoever to Article 13.

Cassie Fenetre says:

Re: So every system must automatically bar false inputs?

And of course it’s another version of Masnick’s reverse sequitur of "corporations are incompetent, therefore must be left entirely unregulated". It’s one of his favorite trick. But of course in reality, if his assertion is true, then regulation is necessary.

James Burkhardt (profile) says:

Re: So every system must automatically bar false inputs?

Except copyright ID is not a DMCA system. It in fact goes much further than the DMCA at the insistence of the rights holders who are advocating for Article 13 (but only if it rejects Safe Harbors). It in fact looks a lot like the system one would require under Article 13. It operates on a different, more stringent framework than DMCA-based red-flag knowledge claims.

The Article 13 filters are “notice and stay down” – meaning you could not restore the content until the content is proven, probably by a court, that the content is not infringement. So your solution fails when Article 13 is brought into the equation. It was mentioned in the article. Did you miss that?

However, your decision to narrow the discussion is disingenuous. Mike is highlighting the fundamental flaws in automatic filters that Content ID demonstrates. Then he draws the clear line to Article 13 and its requirements to prevent uploads, and that the issues present in Content ID, which has had $100 million invested in it, are going to be much worse when the law requires Content ID level filters and places financial and criminal penalties on failures to catch infringing content. And that sites that can’t afford Content ID’s price tag will be working with even worse filters. That is a reasonable connection between 2 current news stories, even if it makes your position look bad.

The article isn’t about resolving the current situation with an unapplicable law. The Article is about questioning how to resolving this same situation under Article 13.

Also, I might note there was a good article a while back about how, at the scale of youtube, a 95% effective filter would still let bad content through (False Negatives) while blocking larger swaths of perfectly fine content (False Positives). (I think it was actually talking porn filters). It gives you a good sense of the numbers you were asking for. I couldn’t find it in my quick search though.

James Burkhardt (profile) says:

Re: Re: Re: So every system must automatically bar false inputs?

In theory, yes that could happen. Most Content ID systems probably weight the thousand Pound gorilla over the Indy artist, so its unlikely.

That said, Jim Sterling makes use of the Copyright Deadlock to prevent ads from showing on his ad free channel. He will intentionally use content from multiple rights holders to get multiple claims, preventing monetization entirely. (He once got Nintendo of America and Nintendo of Japan to deadlock the system by both claiming the same video).

As a note, Jim Sterling provides commentary on the game industry and video games, which requires the use of imagery and footage of game play. He captures his own footage, or provides attribution when another gives him game play. His channel was highlighted by YouTube as the textbook example of Fair Use back when YouTube was making a big show of caring about it, for those interested.

That One Guy (profile) says:

Re: Re: Re:2 So every system must automatically bar false inp

That said, Jim Sterling makes use of the Copyright Deadlock to prevent ads from showing on his ad free channel. He will intentionally use content from multiple rights holders to get multiple claims, preventing monetization entirely. (He once got Nintendo of America and Nintendo of Japan to deadlock the system by both claiming the same video).

As funny as that technique is, I can’t help but find it just a wee bit absurd that he’s essentially forced to game the system in order to prevent others from abusing it.

Anonymous Coward says:

Re: So every system must automatically bar false inputs?

"what percentage of claims are bogus as compared to the [daily millions of?] instance in which copyright works just fine."

That’s a good point. Youtube obviously would have the statistics that would answer this question. However, YouTube seems to withhold a number of key statistics, such as the number of videos uploaded daily (downloads per day are listed in both numbers and total hours, as are total hours uploaded per day). At around .6 million hours of video uploaded daily (not sure if this figure includes livestreams) the number of daily uploads might be somewhere around 1-5 million. Many of these videos are obviosly bot-generated, such as the ones that convert newspaper articles to speech that are spewed out in rapid succession by many different Youtube accounts. Youtube apparently does not filter out these copyright-infringing videos, despite having all the in-house capability to easily do so.

As a multi-billion dollar business, Youtube almost certainly practices some degree of active quality control, as as such sets an acceptable "error" rate of both false-positive and false-negative in its ContentID system. However, it shouldn’t be hard to guess which one takes priority.

Stephen T. Stone (profile) says:

Re:

Exactly NONE of what you example

Example is a noun, not a verb. [This message was brought to you by the Grammar Police: Upholding truth, justice, and the Oxford Comma.]

is Youtube’s fault nor show its ID system fail!

Any number of bogus DMCA claims show exactly how the ID system fails: If the system cannot distinguish between legal and illegal content—or even legal and illegal uses of content, in line with the principles of Fair Use—how “successful” can the system ultimately look from the outside?

To FIX this, Youtube needs only to follow the DMCA and restore the content when contested. Then it’s up to the claimant to continue.

They already do that. (I beat a DMCA once before, so I speak from experience.) The issue lies in the DMCA’s “notice and stay down” system—one I bet you would rage against if it were used against you, even though you have no problem supporting it when “copyright holders” (read: international media corporations) use it against regular jackoffs who are not you.

EUIsCorrect says:

Your Missing the point

I think the point of Article 13 and other measures against Google, Facebook et is that they want a behavioral change from these corporate entities and that they have repeatedly made those points which land on deaf ears.

Fix the problems or face fines that will eventually cripple these corporate entities or force them out of the EU – either method would be a win for the EU and I am with the EU on this.

Algorithms and technology be damned!
If they have to hire half the continent to monitor content and check legitimacy then so be it – the problems these social media companies have created are getting people killed are putting innocent users in legal jeopardy and the ones to blame are companies that store that data and HIDE behind technology. If it’s too big to manage then it shouldn’t be big…

Cassie Fenetre says:

Re: Your Missing the point

You’re new here. Masnick is not missing what you say, he’s muddying up the waters hoping that everyone reasonable will miss it and that Google can go on effectively infringing copyright for its own gain. (And no, I’m not inconsistent with what wrote above: that’s always his goal, and there are many paths to it.)

Anonymous Coward says:

Re: Re: Your Missing the point

“Copyright infringement for its own gain” is silly to try to apply to a search engine. By definition, they exist for the sole purpose of searching for content- it’s like saying that a phone book is infringing on the copyright of a business’s phone number.

(It isn’t, because businesses want you to be able to find their [phone number]/[website URL] so you can contact them and receive their services)

Anonymous Coward says:

Re: Your Missing the point

If it’s too big to manage then it shouldn’t be big…

This is exactly the issue at stake here. The internet is very big, much too big to police in its current state. Now there are fundamentally two ways to deal with that. One is that we develop further in the way we’ve traditionally operated the internet and accept a certain level of collateral damage around IP as a consequence as business models shift and previously closed markets are opened up(both illegally as IP violatons, and legally as publishers become less necessary as a distribution channel). The second is that we convert most of the internet from an open communications channel into a broadcast channel. Most people will have their potential audience cut down to a tiny fraction of what it is now, as only those rich, famous, or influential enough (or who have the support of someone in those categories) will be tenable to platforms, and the throughput of peer-to-peer channels is drastically limited by discovery costs in the absence of search services, to say nothing of the technical challenges of scaling them which are rather beyond the average person (tbh, even setting up a single direct channel is beyond the average person).

JMT (profile) says:

Re: Your Missing the point

"Fix the problems or face fines that will eventually cripple these corporate entities or force them out of the EU – either method would be a win for the EU…"

How would the loss of these companies be a win for the people of the EU? Millions use these products, and even if they want them improved they obviously don’t want them gone.

DoctorMckay (profile) says:

Re: Re: Re: Your Missing the point

I’ve been voting for the pirate party ever since I could vote and will continue to do so.

Now, the elections for the EU parlament are usually not advertised properly thus making so 60% of the population or so does not vote.

In Spain(I live there) we had a 43% turnout in 2014. How sad and manipulated does that number feel?

Scary Devil Monastery (profile) says:

Re: Re: Re: Your Missing the point

"Then they need to tell their elected representatives where their priorities lie. As time goes by it’s getting harder and harder to straddle the fence on these issues."

Copyright was ALWAYS in complete conflict with a great many basic freedoms ensconced in the UN human rights and most constitutional charters in the western hemisphere.

It’s just that lack of viable mass communications technology in the hands of the citizenry has always kept that conflict from erupting. As soon as the tech exists, BOOM. The war erupts for real.

We’ve seen some of this before. The self-playing piano. The Tape cassette and VCR. Media formats like the MP3. Every last new technology has managed to eradicate copyright functionally in the areas it impacted.

And now, we’re in reality up to a point where we either need to abolish the internet – and most functional freedoms – to enforce copyright…
…or we’ll have to accept that copyright was p.o.s. legislation from the start and toss in the garbage pile of history, right on top of "Lese Majeste"

Mason Wheeler (profile) says:

As I’ve said before, the only policy that makes any sense at all in this area is the following:

Infringement is the copyright owner’s problem. It is not my problem, and they have no right to try to make it my problem until after they have proven, in a court of law, that I am part of the problem. If that makes things more difficult for them… not my problem.

Anonymous Coward says:

Re: Re:

That’s the way it used to be, or so I was led to believe.

In addition, copyright infringement is a civil matter. Tax dollars should not be used in pursuit of corporate/private profit. These corporations need to learn a little personal responsibility and stop demanding others make their business work for them.
Copyright infringement is not a felony regardless of what these idiots spew in support of the ridiculous proposed laws attempting to make it so. If your business model relies upon the government to do some of your business activity then said model is flawed and will eventually fail.

James Burkhardt (profile) says:

Re: Section 512(f) of DMCA

One, Copyright ID is not a DMCA system, and does not operate under the DMCA legal framework. 512(f) only applies the DMCA Copyright notices. Does not apply to these cases. That said….

There were a flurry of lawsuits a few years ago, but unfortunately the Dancing Baby Lawsuit effectively but the nail in the coffin of that approach. The bar set by the courts over several rulings was so high that it could never be proven the copyright holder acted in a manner that triggered 512(f) penalties.

Anonymous Coward says:

Re: Re: Section 512(f) of DMCA

(Not a lawyer, but have worked for a website provider handling DMCA notices.)

The DMCA provides platforms with protection from lawsuits from copyright holders if the platform removes content in accordance with DMCA requirements. It also protects the platforms from lawsuits from the platform users whose content is removed if DMCA processes are followed. If ContentID is not a DMCA system, YouTube doesn’t benefit from DMCA safe harbors for videos removed by ContentID and is potentially vulnerable to legal action from all of the users whose material is improperly removed.

While it might be crazy to take on Google, it seems there’s potential for several lawsuits and possibly a class action.

Anonymous Coward says:

Re: Section 512(f) of DMCA

Because there isn’t any regulation or oversight of the DMCA, so you have to counter sue in order to punish someone. That takes time, money, and effort. Which honestly it would probably be easy to punish, or at least get a good enough settlement, some companies who are abusing the system but rarely does anyone try to come together and pool resources to fight back.

They just get the problem fixed and then complain about it via twitter until drowned out by the next person… to get it fixed and just complain about it on twitter.

Anonymous Coward says:

so Now you have safe harbours in the eu,
you won,t be sued by large music companys if you
take down infringing content .
or they may sue the user/uploader .
Article 13 makes youtube legally responsible for all
content uploaded ,
so the record companys will have more leverage against
youtube , give us more money or we will sue you
1000 times over .
The legacy companys don,t care if milliions of users
are blocked from youtube ,
and user free speech is diminished .
ALso the record companys will have less competition from small independent artists ,singers
.
under article 13 youtube may simply block
ALL music content from small creators unless
they can prove they have a license to use the music .
Right now no website has a list of all the songs,book,s, videos ever made so its impossible to make a filter thats would work properly in accordance with section 13 .
Youtube should not just be giving all ad revenue
to companys because they sent them a dmca notice,
they could hold onto the money until the claim is resolved .

Anonymous Coward says:

ahhhh

poor widdle “regulate all the things” crowd is whining and moaning again as regulations fail them.

You will never learn and will always suffer. “it is better to tend to the inconveniences of too much liberty than to deal with attending to too little of it”

“Give me power to protect you and I will only use it to put my boot upon the backs of your necks.”

~Government (no matter which form I take)

Anonymous Coward says:

Takedowns of private videos

On copyright takedowns of non-public videos: the University of California, Berkeley, hosts course lecture recordings on YouTube. Since 2015, the videos are not public and can only be viewed after logging in with a berkeley.edu account. Despite that, several of the videos have disappeared due to copyright claims. (Ctrl-F for "copyright".)

https://archiveteam.org/index.php?title=UC_Berkeley_Course_Captures/YouTube_Status_Fall_2015

  • This video contains content from UMG_MK, who has blocked it in your country on copyright grounds.
  • This video contains content from UMG, who has blocked it on copyright grounds.

https://archiveteam.org/index.php?title=UC_Berkeley_Course_Captures/YouTube_Status_Fall_2016

  • This video contains content from Crowley Media, who has blocked it in your country on copyright grounds.

https://archiveteam.org/index.php?title=UC_Berkeley_Course_Captures/YouTube_Status_Spring_2017

  • This video contains content from BBC Worldwide, who has blocked it on copyright grounds.
  • This video contains content from BBC Worldwide, who has blocked it on copyright grounds.
  • This video contains content from Mosfilm, who has blocked it on copyright grounds.
  • This video contains content from WBTV, who has blocked it on copyright grounds.

As time passes, even more videos from the past disappear. YouTube is not a trustworthy repository for archival videos like these.

Anonymous Coward says:

Re: Takedowns of private videos

These false copyright claims are also the very thing that routinely exterminates dormant Youtube accounts, especially ones with many videos. These accounts will sooner or later “strike out” if no one is there to regularly log in and actively dispute these false copyright claims that hit virtually everyone.

Anonymous Coward says:

the EU politician who wants this in is doing so because he has been ‘encouraged’ to do so by, you guessed it, the same industries that have been after controlling the net for the last 25 years and are extremely close to achieving it. the EU also doesn’t care whether ‘Article 13’ fails or not. all it is interested in is getting it in place so control is achieved and the same industries can take over deciding who can do what on the ‘net and what companies are punished, as the majority are going to be, a lot to the extent of closing down!

That One Guy (profile) says:

Re: Re: Re:

I’m curious, do you find burying your head in the sand a productive way to accomplish your goals? Pretending that something will of course never happen when it’s just shy of happening is a great way to ensure that it does, to the point that I almost have to wonder if you’re in fact in favor of stuff like Article 13 and trying to get people to not take it as seriously as they should.

Wietze Brandsma (profile) says:

Monopsony power

> They just want YouTube to hand them a giant check to “license” their entire catalog. That’s the real endgame here. That it would block out independents and amateurs (i.e., competitors) is just icing on the cake.

Blocking out competitors will give then (more) monopsony power, so they can lower compensation for creators. It looks like it is the other way around.

CharlesGrossman (profile) says:

I love ContentID because I get free money

I uploaded a song to YouTube Music. ContentID told me that a dozen videos use my song, and I get a few pennies every time someone views one of the videos. I’ve looked at all of the videos, and there is NO PART of my song in them. I don’t even know whom to report this to, so I’ll continue collecting my pennies thanks to ContentID.

That One Guy (profile) says:

Nice platform you got there, be a shame if something happened...

The problem is that the ones pushing for this legislative train-wreck would look at the examples you listed and see that as a benefit, rather than examples of why it’s such a problem.

Smaller channels getting slammed by bogus claims, requirements set stupidly high such that sites are much more likely to kill off user submitted content entirely, problems that can be avoided entirely if content producers just sign with the ‘right’ people and hand over the rights to their works…

Or put another way: Stuff like this is a feature, not a bug, to the ones pushing it.

Anonymous Coward says:

Re: Nice platform you got there, be a shame if something happene

problems that can be avoided entirely if content producers just sign with the ‘right’ people and hand over the rights to their works…

Well for the fraction of a percent of creators whose works they accept, the rest can pay an agent to keep on trying until they go broke, or give up trying to get published.

gerg says:

“In short, the $100 million ContentID system is a total fucking mess.”

Even this statement is putting it mildly. Automated AI based filtering will never be able to function properly without some kind of false positive detections ruining things for users. No matter how much money they put into it, it’s impossible to design an automated system that can account for and accurately determine the ways in which content is being used including for fair use, etc.

Urash Zurvan says:

Content ID finds nothing, yet finds something, in the same Video

Because those are so frustrating especially with game videos I now upload test videos to check if I need to mute potions prior uploading, because the Youtube removing or muting thing isn’t working correctly or also takes years.

But I now encountered a new low, I uploaded a 5 Min. Video potion i wanted to test for Content ID claims, after uploading, processing there was nothing, even after reaching HD there was nothing.

So I finished the 99 minute Video where that 5 min. potion also was a part of, and guess what happens after processing it, yeah Content ID claim just in that 5 minute potion that i uploaded hours ago to test for.

So you know, you can’t even test it like this, because it sometimes find something and sometimes not in the exact same part.

Well, so I deleted and re-uploaded that 19GB Video with the claimed part muted, which is way faster than using the Youtube muting tool that needs a day for it while i need 5 minutes.

And in case there’s again a Content ID claim then (who knows with a borked system like this), I will simply delete, mute and re-upoad it again, and again, and again, and again,… Till there is finally peace!

Urash Zurvan says:

Here another method to prevent ads from showing on channels

What Jim Sterling does is interesting, but there’s even a simpler method -> Set the Video to 18+ Adults only, that’s it, 100% no ads and no monetisation possible.

I did that on Videos of Zelda game play that had multiple Nintendo Content ID claims.

And nice side-effect, after a about 8 months those claims suddenly disappeared, who knows perhaps age restricted Zelda Videos send out wrong signals, like if there something obscene in those game or so. But I now still won’t remove that age restriction on them, because they should have thought it over before.

Leave a Reply to JMT Cancel reply

Your email address will not be published. Required fields are marked *

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Ctrl-Alt-Speech

A weekly news podcast from
Mike Masnick & Ben Whitelaw

Subscribe now to Ctrl-Alt-Speech »
Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...
Loading...