Earlier this week one of TorrentFreak’s articles was censored by Google on behalf of a copyright holder.
The article in question was mysteriously flagged as being infringing by an automated DMCA takedown tool. An honest mistake according to the people who sent the notice, but one that doesn’t stand in isolation.
Google previously noted that that 37% of all DMCA notices they receive are not valid copyright claims.
One of the problems is that many rightsholders use completely automated systems to inform Google and other service providers of infringements. They swear under penalty of perjury that the notices are correct, but this is often an outright lie.
Microsoft, for example, has sent Google dozens of notices about the massive infringements that occur on the site Youhavedownloaded.com, a site that is completely non-infringing. As a result, many pages of the website have been de-listed from Google’s search results, directly damaging the site’s owners.
Other rightsholders make even stranger mistakes by massively taking down content that they don’t own. The adult content outfit AFS Media for example asked Google to remove links to the movies Braveheart, Monsters Inc, Green Lantern and many more titles that have nothing to do with the content they produce.
And then there’s YouTube’s content-ID system. We previously outlined many mistakes that were made by the DMCA-style anti-piracy filter, resulting in tens of thousands of ridiculously inaccurate claims.
This week yet another example came up when YouTube labeled birds tweeting in the background of a video as copyrighted music. Again a mistake, but one that probably would have never been corrected if Reddit and Hacker News hadn’t picked it up.
Aside from the mistakes outlined above, there’s also a darker side to DMCA abuse. Google previously revealed that 57% of all the DMCA notices they receive come from companies targeting competitors.
The “competition” angle also ties into the row between Megaupload and Universal Music Group. The latter removed a promo video from the cyberlocker from YouTube on copyright grounds, without owning the rights to any of the material.
It’s safe to say that the DMCA is broadly abused. Thousands of automated notices with hundreds of links each are sent out on a daily basis, turning it into a broad censorship tool. Only the tip of the iceberg is visible to the public thanks to companies like Google who publish some of the notices online.
We can only wonder what’s happening behind the scenes at other sites, but it’s not going to be any better.
Just a few months ago the cyberlocker service Hotfile sued Warner Bros. for DMCA abuse. In the suit Hotfile accuses the movie studio of systematically abusing its anti-piracy tool by taking down hundreds of titles they don’t hold the copyrights to, including open source software.
While we’re the first to admit that copyright holders need tools to protect their work from being infringed, mistakes and abuse as outlined above shouldn’t go unpunished. The DMCA was never intended to be an overbroad and automated piracy filter in the first place.
The above also illustrates why it’s dangerous to allow rightsholders to take entire websites offline, as the SOPA and PIPA bills would allow. The MPAA and RIAA have said many times that legitimate sites would never be affected, but didn’t they say exactly the same about the DMCA?