nav search
Data Center Software Security Transformation DevOps Business Personal Tech Science Emergent Tech Bootnotes BOFH

Google-funded study concludes: Make DMCA even more Google-friendly

You need fewer digital rights. We need more

By Andrew Orlowski, 2 Apr 2016

Comment The DMCA must have been a good idea in its day – almost every other country copied it.

Today, the DMCA's takedown process is broken, but Silicon Valley's billionaire plantation owners have successfully convinced many that it's broken for completely different reasons. Instead of empowering the little guy to give them more ownership over their own stuff, Google argues, we need to punish them even more.

Call it a "cuttlefish strategy." With a cuttlefish strategy, you spray ink into the water so no one can really see what's going on, and Google is very adept at doing this. The ink-sprayers this week are the Google-funded Takedown Project.

Before addressing why it's broken, we need to understand the context in which it was created. The takedown framework was a sop to the individual, in return for which fledgling internet companies would be blessed with an exception from copyright infringement claims.

It made some sense at the time: copyright owners (it's hard to believe today) were once enormously powerful, while internet companies were fledglings. The balance struck allowed the ordinary Joe to remove material from intermediaries without having to consult a lawyer; a one-click, five-minute procedure.

Back then, there were no content ID robots to help automate the business of identifying material, and the burden of checking every envelope and package passing through the postal system would have been so onerous, nobody would have ever wanted to start a better post office.

But power has shifted radically over 20 years. The power now lies with the vast Silicon Valley plantation owners, who have aggregated wealth by ensuring you don't have any control over your own digital goods.

It's not surprising that this super-elite don't want to hand anything resembling "ownership" back to us. Their fortunes have come from monetizing an individual's work – amateur and professional – and selling ads against it. Whether it's Instagram "monetizing" your photos, or Google "monetizing" music on YouTube, it's a business that has worked at scale, and with the minimum of effort and engagement from Google or Facebook.

I doubt that in history, any wealthy elite has acquired quite so much, by doing quite so little. It's actually the people who generate the images, the songs, who do the work. Us, in other words.

The individual in the "digital age" is not just a lobster that's been boiled slowly, oblivious to its imminent fate – but a lobster that has been conditioned to applaud when it sees the chef approaching. Efforts like the Takedown Project, the doxxing site Chilling Effects, and the decade-long "Dancing Baby case" (aka Lenz vs UMG) have something in common other than being funded by the tech industry. (All three receive Google money.) They're all designed to persuade us the tech industry is persecuted by vicious and much-powerful media companies.

The reality is that Google and Facebook are gigantic media companies: the media isn't theirs, it's yours. But they don't want you to know that they're now the world's dominant media companies. You must keep fighting the enemy from 20 years ago.

So these giants will back any cause that makes it harder for the individual to assert their legal rights against a Google or a Facebook, to ensure they have media to sell ads against. The remarkable thing is that some (admittedly on the fringes of the internet, but not uncommon on tech blogs) cheer when the little guy loses, and the big guy wins.

(As an aside, last week Chilling Effect rebranded itself, perhaps belatedly realizing that it successfully fulfilled the very definition of a "Chilling Effect" on society all by itself: "the inhibition or discouragement of the legitimate exercise of natural and legal rights." It's now called something fluffy and cuddly, "Lumen," but it remains in the doxxing business.)

What the DCMA failed to foresee was the rise of broadband and UGC, user-generated content. The law was never intended to provide a shield for unethical behaviour, but that's just how it came to be used.

If you've ever posted a photo online, and it's been washed around the internet without your permission, you'll have had a glimpse of how broken the DMCA really is. In theory, the DMCA takedown process is there to help you. But in reality you're entering a Kafka-esque world taking dozens of procedural steps, followed by shaming and punishment.

How many steps? 46. And you have to plow through some bogus misrepresentations of your rights, designed to put you off. Then, the shaming begins. One of the chief shamers is another Google-funded affair, Chilling Effects, a kind of doxxing project. No matter if you were in the right, if you dared complain, your personal information would be posted for the world to see. Along with details about how to get your stuff.

UGC is the principal loophole on which Silicon Valley is created, where it's most vulnerable, and where it fights the hardest. UGC was supposed to cover storage lockers, not provide a means for a Google to become a global media company without taking any effort to license or curate the material. YouTube still feels like a giant garbage dump, because there's no incentive for Google to take it seriously. As soon as media disappears it pops up again.

Why are there so many takedown requests? Because Google likes it that way. Google could reduce them to zero almost overnight. Google can actually identify every song posted on YouTube, and Instagram can identify almost every copyrighted image. Google's Content ID filtering is mature, being almost a decade old.

However, Google leaves the filters off – unless you sign up to Google's advertising program. Too bad if you don't want to sign up to Google's ad scheme, but somebody else's ad scheme. And too bad if you simply want to assert your rights over your own stuff. Google's strategy more closely resembles that vintage "business model": the protection racket.

(A YouTube employee spilled the beans last year by explaining the ruse in great detail, to cellist Zoe Keating. Without a label, she owned her own recordings. The employee didn't realize Keating was recording the conversation. )

The DMCA's UGC loophole, and the sheer difficulty of doing effective takedowns, ensures Google can maintain an illegal supply chain, and so doesn't need to be too fair when it negotiates its legal supply chain. So guess what the pay rate is in Google's legal supply chain? It isn't very high.

This week's offering from the Takedown Project (Jennifer M. Urban, Joe Karaganis, and Brianna L. Schofield's report, Notice and Takedown in Everyday Practice) is, like many pieces of Google-funded academic research, a lobbying effort lightly disguised as a dispassionate intellectual exercise. (Find it here.) Its authors purport to show that Google is in a world of pain – terrible pain – because of this unfair law. Therefore the law must be tweaked to make it even more Google-friendly, and more hostile to individuals seeking redress.

The study extrapolates that over a quarter (28.4) of DMCA notices are suspect, but this claim begins to fall apart on examination. It's actually a procedural issue: only one in 25 takedown requests wrongly identifies the material. In other words, they're 96.8 per cent accurate, which is not bad for robots.

The authors also admit that takedowns are largely a problem of the online service providers own making, and already possess the tools to solve the bogus "crisis," explaining that "Among OSPs that use filtering technologies, a form of staydown can be achieved – at least for exact or "fingerprinted" matches – when rightsholders put rules in place that prevent any uploads to an OSP's site of new files matching source files."

But they then give away where their sympathies lie. It would be awful and onerous for them do so. They include an anomaly, a European model who has accounts for over 50 per cent of Google Image takedown requests. As indie filmmaker Ellen Seidler points out, in her analysis of the study, "why did researchers include this outlier's data in their results?"

TorrentFreak reports that remarkably, people are flooding the Copyright Office's servers on deadline day to complain how unfair the DMCA takedown procedure is for Google. One day their kids might need the protection over their photos that the law gives them today, and they'll wonder how stoned their parents must have been to demand that those rights disappear.

"Back when you were fighting for Internet Freedom, just what were you smoking, Pop?" ®

*On the fringes of the internet, it's still possible to find people who have not conceived a new way of describing the world, or perhaps not even had a new idea, in twenty years. They remain utterly convinced that newspaper publishers and music labels are vast titans ("Big Media" is usually the giveaway), while Facebook and Google are fragile, tiny startups.

It's a bizarre position to maintain when newspapers and magazines have been folding for a decade, and when Google exerts far more global control over a musician's work than any label ever could. The utopian promise of discovering the internet in 1995 must have been so powerful. But the internet ceased being like it was in 1995 a long time ago.

Sign up to our Newsletter

Get IT in your inbox daily

The Register - Independent news, views and opinion for the tech sector. Part of Situation Publishing