Details Of The Rapidshare Decision: No Search Engine, Plus Efforts To Takedown Copyright Material Helped

from the but-it's-not-a-complete-win dept

Earlier, we wrote about the ruling against Perfect 10 in its attempt to get a temporary restraining order on Rapidshare. As we mentioned at the time, the only details we had were based on Rapidshare’s press release, that obviously presented it in the best possible light to Rapidshare. Thankfully, Christopher Harbin alerted us in the comments that he’s posted a copy of the ruling and written up his own analysis. First, here’s the ruling:

From the initial statements about the ruling from Rapidshare, saying that it could not be infringing since it was just the host, I had naturally assumed that the court found that Rapidshare was protected by the DMCA’s safe harbors. But that is not the case. In fact, the court basically tosses out Rapidshare’s safe harbor claim, because Rapidshare (bizarrely) has never set up an official DMCA agent — which is necessary to qualify for the safe harbors. I would guess that Rapidshare’s excuse for not doing so is that it’s not based in the US, but if it’s going to deal with lawsuits in the US, it’s crazy not to have set up a DMCA agent (a very simple process).

That said, there are some interesting things (not all good) in the ruling. There were a few things that helped Rapidshare here. First, is that it does do things to takedown content it believes to be infringing. Second (and perhaps most interesting) is the fact that it does not index or promote the content stored on its service by itself. Since most file sharing services include a search engine aspect, the court concludes that this is a key element of suggesting that Rapidshare is not liable for direct infringement. That’s an interesting ruling that could answer the question I’ve asked before about whether or not it actually is possible to create a site that allows for sharing of files that does not run afoul of the Grokster inducement rules.

The other thing that helps Rapidshare is that, despite being involved in a whole bunch of copyright cases, it still appears that Perfect 10 is somewhat incompetent in understanding copyright law and availing itself of tools to deal with infringement. In this case, Perfect 10 didn’t send Rapidshare the specific links where infringing content existed. It just sent the company a bunch of files, but Rapidshare had no way to match those files to ones hosted in users’ lockers. Even then Rapidshare still did try to find as many of the images as possible, even going so far as to do Google searches, and delete them. On top of that, Rapidshare has a tool that lets content owners indicate if certain files are infringing. Perfect 10 chose not to use that.

The one part of the ruling that I found troubling, however, is that the court did say that, even without being alerted to what’s infringing, Rapidshare did have “specific knowledge” of infringing works on the site. This is a key point of dispute in the Viacom/YouTube case, where Google makes the argument that there’s no way for it to know which works are really infringing and which are there on purpose. Yet, in this case, the court says that the service provider does have specific knowledge:

Thus, it appears that specific knowledge of direct infringement may exist even where an operator does not have information that would allow it to search its contents and distinguish infringing from non-infringing materials. Here, RapidShare received notice of hundreds of copyrighted Perfect 10 images that were found on its servers. The Court therefore concludes that RapidShare had actual, specific knowledge of direct infringement.

Basically, it says that because Perfect 10 alerted Rapidshare to infringing content, then Rapidshare has “specific knowledge.” That wasn’t enough to meet the full inducement standard, so isn’t a huge issue here, but that logic seems wrong to me. If that’s all it takes to create “specific knowledge,” then lots of user-generated content sites could be in trouble. So, if you want to cause trouble for any UGC site, you just upload content, and then alert the site that there is infringing content on the site, but don’t tell them where it is or how to find it. I don’t see how that’s “specific knowledge” at all, but the court felt otherwise.

Finally, while I appreciate Christopher’s analysis of the ruling, I have to disagree with his first two points. He doesn’t understand why not having a search engine matters:

Although Rapidshare does not index its files, it basically punts indexing to third-party websites. It?s trivial to find infringing material hosted on Rapidshare and other file-hosting sites and I?m not sure why dicing up storage and indexing into separate entities which obviously have a symbiotic relationship should be able to avoid liability.

I think that’s a slight misread of the ruling. It’s not saying that Rapidshare gets to avoid all liability. It’s just that it takes away Rapidshare’s direct infringement liability, because it is not, in fact, setup to make it easy for people to find that content. The fact that others have made it easier to find that content should not be blamed on Rapidshare.

Second, Christopher notes:

I’m not at all convinced that in all cases plaintiffs should be forced to ferret out all infringement on a defendant’s website. As the law stands right now, copyright holders have to employ an army of people to constantly monitor defendant’s site for infringement.

To which the obvious response is, why should that burden fall on the defendants? If it takes plaintiffs “an army of people,” it would actually take the defendants significantly more than that, complicated by the fact that the defendant has no way of knowing if the content has been uploaded legitimately or is an infringing copy.

Filed Under: copryight, dmca, infringement, liability
Companies: perfect 10, rapidshare