Yesterday I responded to Solveig Singleton’s comments about the Linux DVD player issue. Today I want to focus on the other major argument of her posts, namely that DRM is sufficiently effective that the DMCA is worth it even if it does have some disadvantages.
Unfortunately, the debate over whether DRM is effective often has a “Does not! Does too!” quality to it. I’m going to try to dig into the matter a little more deeply to see if we can get past that to some substantive discussion of when DRM is effective and when it isn’t. I apologize in advance for the length of this post.
To start off, here’s a stylized version of the typical debate over the effectiveness of DRM, in which I’ll cast Jack Valenti as the pro-DRM advocate:
JV: Our industry is being overwhelmed by piracy! Millions of young people are getting music and movies for free that they ought to be paying for. We need new technological and legal tools to protect our rights.
TL: This is nothing new. Since at least the 1970s, every new media technology–the cassette tape, the VCR, CD burners–has brought with it new capabilities for copying. The RIAA and MPAA has cried wolf with each technology, yet in each case the warnings proved overblown. The movie industry seems to have survived the tape deck, VCR, and CD burner just fine.
JV: But this time it’s different, because the Internet changes everything. Peer-to-peer applications allow people, with the click of the mouse, and without authorization, to send illicit files hurtling at the speed of light to every nook and cranny of this planet. Piracy tools are a lot more efficient than they used to be, and so we need new tools to fight back.
TL: But as the darknet critique shows, DRM isn’t going to prevent the spread of illicit files on peer-to-peer networks. All DRM schemes have flaws, and it just takes one person to crack the DRM scheme and upload the file to make the DRM scheme completely ineffective.
JV: Well sure, nobody said that DRM was perfect. Obviously, sophisticated hackers will always be able to break DRM if they try hard enough. But at least DRM will have some deterrent effect. It will keep honest users honest, and reduce the temptation of busy users to infringe.
There are actually two very different types of copyright infringement being discussed here, and I think a lot of sloppy thinking results when we confuse them.
The first, which I’ll call meatspace infringement occurs when you copy an orginal copyrighted file from someone you know in real life. This might include using your two-cassette tape deck to copy an audio tape, burning a copy of your friend’s CD, or running DeCSS on a friend’s DVD to make a copy for your laptop.
The second, which I’ll call cyberspace infringement occurs when you receive a file from a stranger on the Internet. This typically occurs using a peer-to-peer application like Grokster, but can also occur on the web, in a news group, or via email.
If you look at my imaginary exchange with Mr. Valenti, you’ll see that he’s promiscuously mixing up these two types of infringement. Let’s try to pull them apart and see how effective DRM is against each.
Meatspace infringement: DRM indisputably does deter some meatspace infringement. It’s easier to make a copy of a CD than it is to make a copy of a DVD, because the latter has CSS. I have to go out to the Internet and download a tool like DeCSS if I want to make my friend a copy of my DVD. However, the question is why is this any different from past generations of copying technologies? Banning dual-tape decks, VCRs, and CD burners would also reduce piracy. Yet people generally agree that doing so wouldn’t be worth the cost. They recognize that, as the Supreme Court put it in the Betamax decision, we need to balance the need for effective copyright protection against the “rights of others freely to engage in substantially unrelated areas of commerce.” Hollywood’s ownership of (some) copyrighted TV shows doesn’t give it the right to dictate what features VCRs are allowed to have. By the same token, Hollywood’s ownership of (some) DVD movies shouldn’t give it the power to decide who’s allowed to build a DVD player.
So why is 2006 different from 1984? That brings us to:
Cyberspace infringement: This really is new. It’s a big problem that (as far as I’ve seen) no one has really proposed a satisfactory solution to. It didn’t exist in 1984. But here’s the thing: DRM doesn’t prevent cyberspace infringement. It doesn’t even slow it down very much. As soon as one person cracks the DRM and uploads a copy to a peer-to-peer network, DRM has become useless. And on an Internet with a billion users, the cracking process is likely to measured in days, if not hours or minutes. It’s not much of an exaggeration to say that DRM has no effect at all on the prevelance of cyberspace infringement.
When I wrote that DRM doesn’t prevent piracy, this is what I was talking about. I used the analogy of the Maginot line in my paper: no matter how high you make the DRM walls, it only takes a single breach and DRM becomes completely ineffective against cyberspace infringement, as everyone “drives around” the wall by downloading an already-cracked copy from a peer-to-peer network.
So with that very long introduction as background, let’s turn to Singleton’s argument:
DRM does not prevent all piracy. DRM schemes can be broken, some rather quickly, by determined hacktivists or whoever. The results, both hacker tools and decrypted content, are readily available on the Internet. But you have to go get them. And they are not sanitized in the manner of licensed offerings. An awful lot of consumers just don’t have the time or inclination to mess with this sort of thing.
I’m not sure I follow her example, as it sounds like in both cases he’s violating the DMCA (region-free DVD players are presumably circumvention devices) But in any event, she’s clearly talking about meatspace infringement here, and correctly noting that DRM has some deterrent effect on it. But then in the very next paragraph she switches to talking about cyberspace piracy:
DRM does not stop determined pirates. But that, coupled with the DMCA, makes it necessary for those making unauthorized copies to use black-market sources that are not especially convenient. This is enough to stimy those who would free-ride if the cost in time and trouble were near zero, but not otherwise. The idea that there are two classes of content consumer, the determined pirate and the honest consumer, and nothing in between, is nonsense. There are in my experience precious few in the latter category, especially in certain age groups. My young brother-in-law, for example, reported that he was the only student at his university he knew who did not download music from unlicensed P2P networks.
The cost of meatspace infringement is never “near zero.” Even back in the days before DRM, when dual-tape decks and pre-macrovision VCRs were the norm, lots of people paid for music and movies. The combination of convenience and a clean conscience was worth the money to them. The primary barrier to meatspace-copying a DVD isn’t CSS, it’s the time (and possibly the embarrassment) required to find a friend who has the DVD you want, borrow it, purchase DVD-R media, and make the copy. Most consumers wouldn’t bother to do that even with no CSS, just as most people who own CD burners still buy CDs. If that weren’t the case, the recording industry would have gone out of business in the 1970s, and Hollywood would have gone under in the 1980s.
But in this paragraph, she’s suddenly switched to talking about cyberspace infringement. The cost of that has been dropping rapidly, and at least for college students with a lot of time on their hands, its cost is pretty close to zero. But you’ll notice she doesn’t give any example of cyberspace infringers who are deterred by DRM. DRM simply doesn’t have any effect on peer-to-peer infringement.
So we seem to have a mild case of schizophrenia here. Singleton claims (1) that DRM reduces infringement, and (2) that unchecked infringement is destroying the content industries. Together, these seem like a pretty compelling case for DRM, at least until you realize that (1) and (2) refer to totally distinct types of infringement. (1) is about meatspace infringement, which isn’t a significantly bigger problem today than it was 10, 20, or 30 years ago. (2) is about cyberspace infringement, against which DRM is totally ineffective.
This reminds me a little bit of the experience I have every time I go into an airport. Airports these days are full of pointless security restrictions. My local airport, for example, has closed the closest couple of rows, presumably to prevent someone from driving a truck bomb into the airport. I have to take my shoes off for scanning. I have to surrender nail files to the security guards. When I point out that this is silly, the usual response is “9/11 changed everything.” Ever since September 11, no proposal to beef up security is too outlandish, even if most of it wouldn’t have done a thing to stop an attack like 9/11.
Similarly, Singleton seems to believe that illicit peer-to-peer networks change everything. And she may even be right. The problem is that the policy solutions she offers won’t do anything to stop illicit peer-to-peer networks. Her policy prescriptions only have an effect on old-fashioned meatspace infringement. But there’s no reason to think meatspace infringement has become a bigger problem in recent years. We didn’t think the threat of meatspace infringement was worth banning copying technologies in the 1970s, 1980s, or 1990s, and there’s no reason we should think differently today. Banning DeCSS to stop Grokster is akin to making people park farther from the airport to prevent another 9/11. It’s not completely useless, but the policy prescription isn’t really connected to the problem it’s purported to solve.