Update: Should there be a rating system for “red flag” knowledge: Are Five Million DMCA notices too many?

[3/31/13 Update by Editor Charlie:  We now know courtesy of Google’s DMCA Transparency Report that Google processes roughly 10,000,000 DMCA notices a month for search alone which Google acknowledges to be 97% accurate.  It also seems likely that the Google lobbyist had some idea or actually knew of this level of infringement in Google search at the time she testified.  And Google sends traffic to pirate sites where it also sells advertising on a CPM basis–profiting from the theft.  When will someone call this out for what it is–a criminal agreement to profit from massive infringement?

12/27/11 Update by Chris Castle:  In written testimony before the House Judiciary Committee on November 16, Google’s lobbyist acknowledged that “During 2010, we processed DMCA takedown notices for approximately three million items across all of our products. Already in 2011 we have processed takedown notices for nearly five million items, and we have done so more quickly and efficiently than ever before.” 

Interesting, because of a couple things:  First, that’s a 166% increase year over year.  Second, in her oral testimony, she left out the 3 million in 2010 part.  Maybe that was just an editorial decision, but if the members had not read her written testimony (highly likely) and didn’t focus on that part, they probably would not–and didn’t–raise the rather startling growth of DMCA notices “processed”.  And finally that word “processed”–how many of you reading that statement thought she meant “taken down”?  If you have had any experience with Google (see www.popuppirates.com) you would know that there is a very, very long way between “processed” and “taken down”.

We were also treated to many recitations about how much money the tech companies make off of the Internet and the DMCA.  The implication is that’s all magically new money.  How much of it is more income transfer than profit from innovation?

The following piece written by Chris Castle first appeared in MTP in 2006.]

In a recent conversation I had with an attorney who has represented YouTube, I asked him a question that I now feel very safe in asking—does a million notice and takedowns sent to one service in 12-18 months constitute “red flag” knowledge? This isn’t speculation anymore–YouTube must have received at least a million DMCA notices by now.  Even if the site has a repeat infringer policy, should they still get a safe harbor if they seem to attract significant numbers of repeat infringers?

Realize that if a million notices constitutes “red flag” knowledge of infringing activity, the recipient infringer may be denied the protection of the safe harbors in the Digital Millennium Copyright Act.

His reply? Well, you see, it depends on how large the service is.

Ah, I see. So if the “service” does a really good job of creating lots and lots of infringement but only gets caught a small percentage of the time, then that doesn’t mean that they “knew or should have known” that infringing activity was going on. Even if that small percentage was over a million copyright owners who decided to send notices. (Realize that at least 10% of that hypothetical million was the Viacom lawsuit alone, and another 5% was JASRAC alone, etc.)

So if your average “citizen of the 21st century” happens to run across their work on YouTube but doesn’t know what to do to take it down, phrases their notice incorrectly–what then?

Or what if the artist “citizen of the 21st century” can’t afford a lawyer “citizen of the 21st century” to tell them whether the use involved is “fair use” or whether the artist’s unschooled consideration of whether the infringer’s use is subject to a “fair use” affirmative defense was accomplished with sufficient “good faith”, yet another new wrinkle courtesy of The Leland Stanford Junior Google Law School lecturer and U.S. District Judge Jeremy Fogel (who essentially codified what we understand to be the DMCA practice of Stanford benefactor Google in his ruling in the Lenz case)?

Nothing will happen. Nothing. And these artists who throw up their hands out of economic desperation will also make their contribution to the $1.65 billion of free rider profits that the YouTubes put in their pocket from the work product of others that they can’t hold a candle to.

So how does the average “citizen of the 21st century” decide how to spend their time monitoring the Internet for infringements of their works? Google and its amen corner continually tell us that was the deal that copyright owners made in the DMCA—catch us if you can. Democracy guided by the hand of the Almighty.  Not true, of course, and bad advice, most certainly, but that’s what a Google might say.

Wouldn’t it be more efficient for an artist “citizen of the 21st century“–who wishes to enjoy all the extraordinary benefits the first decade of the 21st century is bringing to them–if there were some kind of ranking system for the really bad guys? A ranking system for copyright infringers based on DMCA notices sent? (Or at least “potential” copyright infringers, if you want to continue the charade of the wilfully blind.)

What if there were a website that maintained an informal ranking of which online service received the most notices, and also gave time to response, and time to takedown? Kind of like an Alexa.com meets Good Housekeeping for infringers? Which “digital native” amongst us could possibly object to consumer information being made available to “citizens of the 21st century” ? Maybe the site could also post a picture of the DMCA agents who failed to respond, kind of like johns.com? (Maybe conveniently available on Google Street View.)

There’s actually nothing that would stop artists from developing such a service–aside from money, of course, and time taken away from diligently monitoring the Internet for other “citizens of the 21st century” who are infringing their work. You would think that the Copyright Office would want to maintain it, too, so they could see how effectively the DMCA notice and takedown system was working. Or maybe even the Congress might be interested?

In Los Angeles, the city health department gives restaurants a rating card that is posted on the front door of the establishment. If you don’t have an “A” on that card, expect to do some explaining to your customers.

It’s amazing how much of an effect a rating can have on the failure to comply with the law.

3 thoughts on “Update: Should there be a rating system for “red flag” knowledge: Are Five Million DMCA notices too many?

  1. Content ID is supposed to filter out the infringing uses. It’s an alternative to sending a DMCA notice.

    If Content ID is working a significant percentage of the time, then YouTube would probably get an “A” rating. The current state of play with Content ID is that Content ID seems most effective with major label artists and major studio pictures. This is in part because major labels can afford the anti-piracy staff to communicate with YouTube when someone catches a track that got through Content ID. When those tracks are caught (like a pitch bend situation), no notice is sent either (because YouTube takes them down when contacted). If artists that aren’t signed to a major label get this courtesy, it’s kept very quiet because I’ve not heard of it at all.

    If YouTube is still attracting DMCA notices, that would be after the application of Content ID, so that is a fact that artists and songwriters might want to know. The point of the post is that it doesn’t appear that anyone knows how many take down notices are received except for YouTube. If the number of notices is high, then YouTube would probably not want that fact known. If Content ID is working as you suggest, then presumably the number of notices is low which one would think YouTube would be proud of, like the restaurant that gets an A rating.

    It would also be valuable to disclose the number of counternotifications.

    I would also think that the Congress would like to know the raw numbers regardless of how well a filtering system works.

    This, by the way, has nothing whatsoever to do with the proposed “rogue sites” legislation. It has to do with the interpretation of “red flag knowledge” under existing law. The point being if something happens a million times that is a proxy for failure of a filtering system, something’s not working very well, kind of like a red flag.

    Like

Comments are closed.