DMCA Mysteries

Jane Ginsburg of Columbia Law School has a paper called “The Pros and Cons of Strengthening Intellectual Property Protection: Technological Protection Measures and Section 1201 of the US Copyright Act.” The paper paper is a thorough and readable survey of recent legal decisions regarding the DMCA’s anti-circumvention provisions.

The paper highlighted something that I hadn’t given a lot of thought to before: the DMCA ostensibly prohibits two separate kinds of circumvention. Section 1201(a) prohibits circumventing “technological measure that effectively controls access” to a protected work, as well as “trafficking” in devices for that purpose. Section 1201(b), in contrast, prohibits trafficking in devices that circumvent a “technological measure that effectively protects a right of a copyright owner.” Ginsburg discusses this distinction at some length.

It seems to me that the obvious thing to say about this distinction is that it doesn’t make any sense. DRM schemes typically do not attempt prevent copying, what they attempt to do is prevent playback using encryption. You can make as many copies of a DRMed file as you like, but it won’t do you any good unless you have the key required to decrypt it.

It should be obvious, then, that any device that runs afoul of the “rights” prohibition must run afoul of the “access” prohibition in the process. Either you know how to decrypt (and therefore “access”) the file or you don’t. (You might also intercept the content after it’s been decrypted, but this doesn’t change my point) The “circumvention” part of the process is the same regardless of whether you’re just trying to access the file yourself or you’re planning on sharing it with a million of your best friends.

Now, it is conceivable that someone could write a program that ran afoul of the “access” prohibition but not the “rights” prohibition. If, for example, someone wrote a utility that allowed users to play their iTunes songs on a device not supported by FairPlay–but not to convert it to an unencrypted format–that might violate 1201(a) but not 1201(b). But as far as I can see, the reverse–a device that violates 1201(b) but not 1201(a)–is not plausible. What would it mean to create software that allowed you to circumvent copy controls without circumventing access controls?

Which, as far as I can see, makes 1201(b) redundant, and Ginsburg’s discussion of the distinction a little bit silly.

But that’s a nitpick. The paper gets a lot of other things right. She points out that the Copyright Office’s triennial rule-making process is of limited value because that process only gives exceptions for the act of circumvention, not for “trafficking” in the software that most consumers would need to actually take advantage of the exception. And she points out the limits of the reverse engineering exception in 1201(f):

At first blush, the “reverse engineering” exception, §1201(f) might appear on point, given its goal of “achiev[ing] interoperability.” On closer examination, however, the goal of interoperability with other playback devices does not seem to conform to that provision’s conditions. Section 1201(f)(1) states: “Notwithstanding the provisions of subsection (a)(1)(A), a person who has lawfully obtained the right to use a copy of a computer program . . .” A music file is not a computer program. It is in digital format, but not all works expressed in 1s and 0s are “computer programs.” Access to the digital file may be controlled by a computer program, but the subject matter addressed by §1201(f) is the computer program whose access is being controlled, not the computer program controlling access. The exception thus fails at the threshold.

This is quite right. I find the purpose of section 1201(f) to be nearly as mysterious as the distinction between access controls and rights controls. Whatever it was supposed to mean, it pretty clearly doesn’t give people a broad right to circumvent copy controls for the purpose of making their devices compatible.

Given her generally excellent discussions of the limitations of the DMCA’s circumvention exceptions, I found it a little strange that she concluded that she concluded that “To date, ‘digital lock-up’ persists in spectral guise, a grim, yet untranspired, anticipation.” She does not even mention FairPlay, which seems to me the most obvious example of “digital lock-up.” Nor does she discuss the launch of Zune, which is mimicking Apple’s locked down business model–nor the various movie services, most of which lock the user into their particular file formats. Perhaps her point is that there have been no court decisions explicitly ratifying the view that, say, the DMCA prohibits companies from reverse-engineering FairPlay to play iTunes songs on their MP3 players. But there also don’t appear to be any products on the market that do such reverse engineering. (Jon Johansen is offering a library to do so, but so far I don’t think he’s gotten any takers) The threat of a lawsuit is often as effective as actually filing the lawsuit, and so the fact that there hasn’t been much litigation on this front does not prove there hasn’t been a chilling effect.

SHARE: 20 comments