After reading James DeLong’s defense of Patrick Ross’s CNET article on the DMCA, I have to admit that I was being unfair to call the article “incredibly confused.” I interpreted Patrick to be saying that the DMCA literally allows consumers to break contracts they’ve made with content providers. But after reading DeLong’s defense, it’s clear to me that what Patrick meant is that a digital rights management scheme is like a contract in the sense that it allows publishers to place various restrictions on the use of their intellectual property and sell it at different prices. While this isn’t literally a contract (as Patrick might concede) perhaps it’s contract-like device that allows certain beneficial transactions to occur (such as a limited-time online rental of a movie) that would otherwise be impossible.
There’s a kernel of truth to this argument. Certainly if a content producer knows that renting videos online will lead to rampant piracy and destroy the market for purchasing videos, the producer will be reluctant to offer a rental option. Consumers clearly don’t benefit from fewer choices.
But this analysis misses two important points. The first is philosophical: a DRM isn’t a contract, and in several important ways, it’s not even very similar to a contract. As I noted in my previous post, the DRM “contract” is entirely one-sided. Its terms are set by the publisher and the publisher has the power to unilaterally and retroactively change them. If Apple sells a song to a consumer, it is under no obligation to ensure that the DRM scheme enforces the terms stated at the time of sale. Apple is free to change or reduce the functionality available to the consumer (such as reducing the amount of burning permitted) and the consumer has absolutely no recourse.
Secondly, the terms of the DRM “contract” are never clearly disclosed to the purchaser–they certainly aren’t disclosed prior to sale. If you polled Apple’s customers at the time of sale, I bet only a tiny fraction of them would be aware that they were prohibited from listening to the song they just purchased on an MP3 player from Dell, Sony, or other vendors. Similarly, nowhere on the packaging of a DVD does it state that you’re not allowed to play the DVD on the Linux operating system.
Most software licenses have a return clause where, after reviewing the terms of the license, you can return the unopened software for a refund. There is no analogous process for DRM. If a consumer buys a song on iTunes hoping to play it on their Dell device, they have no opportunity to return the song on discovering they don’t have that option.
Thirdly, and most importantly, contracts are government by contract law. Contract law is significantly different from copyright law. The principles of contract law recognize that not every term of a contract should be enforced to the letter if doing so is in the interest of neither party. For example, most concert tickets have a no-resale provision, yet no one thinks anything of selling one’s concert tickets to a co-worker. We recognize that the concert organizer probably doesn’t mind such casual resale (since the ticket would otherwise have gone to waste) and so if the concert organizer doesn’t check IDs at the door, the state shouldn’t get involved.
DRM schemes have no such flexibility. Every provision of the DRM “contract” is enforced to the letter, regardless of whether such enforcement makes any sense. If someone want to use a DRMed song in a way that the creators of the DRM scheme didn’t forsee (say, your daughter is putting together a video for a class project and wants to use a particular song as background music), you’re simply out of luck. It might be that Steve Jobs would be happy to have your daughter use that song, but the DRM scheme has no way of making case-by-case exceptions.
That’s how we get such absurdities as DVDs that can’t play on Linux computers, DRMed songs that can’t play on certain MP3 players, eBooks that don’t allow the use of screen readers, and $20,000 audio setups that can’t play DRMed songs. Because DRM lacks the flexiblity and context-sensitivity of a real contract, it places arbitrary and pointless restrictions on anyone who wants to do something not forseen by the DRM creator.
The second thing this analysis misses is that DRM just isn’t an effective way to prevent piracy. The problem is that it only takes one person to “crack” the DRM, and then everyone who wants an unrestricted copy can get it from that one person. This post is already too long, so I’ll just link to a Microsoft paper that does a good job of explaining the dillemma.
In practice, most consumers don’t want to engage in piracy. To the extent that DRM schemes prevent piracy at all, they do so by putting up “speed bumps” to discourage generally law-abiding folks from casual sharing of copyrighted material. But those “speed bumps” would be just as effective in a world where circumvention is legal. Downloading cracking tools will always be a somewhat seedy and inconvenient process, and so most law-abiding consumers won’t do it even if it’s legal. On the other hand, in cases where the DRM scheme becomes a major impediment to a lawful activity (say someone decides to switch from an iPod to a Dell MP3 player) the availability of “circumvention tools” (such as, say, a Dell-provided software tool that converts iTunes songs to a format suitable for use on Dell’s MP3 player) provides an important safety valve to prevent the DRM scheme from placing unreasonable restrictions on consumers.