by Adam C. Engst <email@example.com>
Much has been written about what's wrong with the Digital Millennium Copyright Act (DMCA). After all, it's been used to jail programmers, threaten professors, and censor publications, and because of it, foreign scientists have avoided traveling to the U.S. and prominent researchers have withheld their work. In a white paper about the unintended consequences of the DMCA, the Electronic Frontier Foundation argues that the DMCA chills free expression and scientific research, jeopardizes fair use, and impedes competition and innovation. In short, this is a law that only the companies who paid for it could love.
Just who are we talking about here? Primarily the large movie studios and record labels, who own the copyrights on vast quantities of content and who have been working with one another and via their industry associations, the Motion Picture Association of America (MPAA) and the Recording Industry Association of America (RIAA), to control how we are allowed to interact with that content. Their unity of purpose and storm-trooper tactics have led some to dub them the "Content Cartel."
However, the DMCA is merely one link in a chain that's being used by the Content Cartel and many others to restrict access to the shared cultural heritage of the world, and in the process, extract money from our pockets, stifle innovation and competition, and protect entrenched interests.
DMCA and Trusted Systems -- I recently attended a talk by Professor Tarleton Gillespie <firstname.lastname@example.org> of Cornell University in which he made a compelling argument for how the Content Cartel is using the legal force of the DMCA to direct us down a path where content cannot exist outside of a "trusted system," which is a set of hardware, software, and file formats that all agree on what the user is allowed to do with a piece of content. (The trust here is between the pieces of the system, because the content owners don't trust their customers at all.) The trusted system's goals are simple - to eliminate all unauthorized uses and create a situation where we pay more for the content we consume.
A trusted system could prevent you not only from copying a CD or DVD, but also from listening to the CD more than a certain number of times in a day or skipping commercials on a DVD or on broadcast television. Along with requiring us to buy new hardware to play such content and buy new protected versions of the content we already own, a trusted system could have another ill effect. That's because it could prevent us from working with content we would create, using tools such as those Apple kindly provides in iMovie, iDVD, iTunes, and iPhoto. In the worst case scenario, Apple could lose not just the Mac's current digital media advantage in the marketplace, but the ability to work with digital media at all. See Cory Doctorow's article on the broadcast flag in TidBITS-642 for more on this disturbing possibility.
Professor Gillespie illustrated how this could happen with a discussion of the awkwardly named Content Scramble System (CSS), used to prevent people from copying DVDs, and the DeCSS software created by a Norwegian teenager with help from others on the Internet to build a Linux DVD player.
(A brief aside: DeCSS violates the DMCA's anti-circumvention provisions, which ban devices or services that are designed primarily to circumvent copy prevention technologies, that have only limited commercially significant purpose other than circumvention, or that are marketed for circumvention. The DMCA was signed into law in large part to bring the U.S. into compliance with a pair of World Intellectual Property Organization (WIPO) treaties that require anti-circumvention protections in the copyright law of signatory nations. You might think Norway would be included among the nations signing these WIPO treaties, but in fact, only 37 countries have signed on, including the U.S. and Japan, along with the likes of Kyrgyzstan, Gabon, and Paraguay. We're not talking about full international support here, especially in contrast to the 149 signatories to the more general and long-standing Berne Convention for the Protection of Literary and Artistic Works.)
In particular, Professor Gillespie focused on three defenses used in the court case filed against Eric Corley, publisher of the hacker magazine 2600, by eight movie studios to prevent 2600 from publishing the DeCSS software. Although Eric Corley didn't create DeCSS, he made it available on the 2600 Web site. His lawyers' defenses focused on ways DeCSS might escape the anti-circumvention provisions in the DMCA, which was the law under which the case was being tried.
Let's look at these defenses, all of which the court eventually dismissed in ruling for the movie studios and enjoining 2600 magazine from posting the DeCSS code. A subsequent appeal also failed, and the defendants chose not to appeal again to the Supreme Court (probably a wise move - this particular case struck me as fairly weak).
Create a Linux Player -- The primary defense that Eric Corley's legal team, funded by the Electronic Frontier Foundation (EFF), advanced was that CSS was reverse engineered and DeCSS written to further the development of a DVD player for Linux, which allegedly had no way of playing DVDs at the time (four players are available now; see the Linux Journal review linked below for details). Unfortunately, the judge deemed the defense utterly irrelevant because the DMCA offers no relief based on motivation. In short, if a technology violates the DMCA's anti-circumvention provisions, the purpose for which that technology was created simply doesn't matter. The judge also wasn't impressed with the fact that DeCSS is actually a Windows program, so although it could be argued that it was a necessary step in the creation of a Linux DVD player, it's a weak argument.
The obstacle that actually lies in the way of creating a DVD player is the lack of a key to decrypt the CSS encryption used on DVDs. The only way to come by such a key is to sign a contract licensing CSS from the DVD Copy Control Association (DVD CCA), a group made up of companies representing the movie studios, consumer electronics companies, and the computer industry. At $15,500, the licensing cost is not usurious, but the contract effectively prevents individuals and small organizations from licensing CSS. For instance, in the event of a material breach of contract, the licensee is liable for $1 million, and damages can grow to a maximum of $8 million. In addition, the contract prevents licensees from reverse engineering CSS or working in any way counter to the goal of CSS's protection of DVDs.
Put simply, the CSS license is the sort of thing only large companies can reasonably sign, so it's clear that the effect of the DVD CCA contract is to keep newcomers out of the cozy little club. Perhaps that wasn't a likely concern before the age of the Internet, but the rise of Linux and the open source movement shows that small, informal groups organized over the Internet can produce software that threatens the largest of companies.
The end result here is that innovation is stifled. Companies that license CSS cannot, even if they wanted to, produce products that consumers might like to buy, such as DVD recorders that could copy a DVD. That keeps new companies, niche players, or even independent programmers from competing with the consumer electronics giants with innovative features that in any way run afoul of CSS. So although the consumer electronics companies might not have minded consumers copying DVDs, since they would sell the equipment to make that happen, it's worthwhile for them to abide by CSS to eliminates potential competition.
Equally as problematic is that the CSS license's numerous requirements force the consumer electronics firms to be technologically responsible for regulating our movie viewing and copying behaviors for the studios. Signing this draconian contract is an all-or-nothing deal, so the movie studios have cleverly managed to pass off the dirty work of technological regulation on everyone else (they just produce the content; the DVD and player manufacturers must implement CSS). It's a big step toward a trusted system in which all the parties are bound by the CSS contract.
(As an aside, another effect of the CSS contracts is also to move the entire issue from the world of copyright law, where there is at least some presumption of needing to benefit the public, into the world of contract law, which doesn't give a damn about the public good. If this continues to the logical extreme, the concept of copyright, and unauthorized access to any content, could be locked up forever in simple contracts that lie underneath a trusted system's technologies, all backed up by the DMCA's anti-circumvention provisions.)
Perform Encryption Research -- Another defense that Eric Corley's lawyers put forth was that DeCSS was created as research into the CSS encryption method, since the DMCA does allow copy-prevention technologies to be circumvented for encryption research. However, the DMCA specifically requires that the encrypted copy be obtained lawfully and that the person performing the research make a good faith effort to obtain authorization in advance. In addition, the decryption tools from such research may be shared only with collaborators for good faith research purposes - in other words, distributing these tools publicly isn't kosher.
Note the words "good faith" above. In determining whether encryption research is good faith, the judge said the court must determine whether the results are disseminated in a way that advances the state of knowledge of encryption technology, whether the person is engaged in legitimate study of work in encryption, and whether the results are communicated to the copyright owner in a timely fashion. Deciding that none of these tests were true of Eric Corley, the judge dismissed out of hand the claims that DeCSS had protection under the encryption research exception to the DMCA.
Looking past the specifics of this case, consider the ways in which encryption research is considered to be in good faith. You must be a legitimate researcher, have a goal of advancing the state of knowledge, and have at least made an effort to get authorization from the copyright owner. Now think about how these requirements completely disenfranchise the interested individuals and the Internet technical geek community. What does it take to be considered a legitimate researcher - a white coat, thick glasses, and a job with a university, corporation, or government body?
What we're seeing here is how the DMCA in essence props up the status quo, denying that legitimate research could be done outside the halls of academia or a company's R&D department. Left on the outside are the crazy ones, the misfits, the rebels, the troublemakers... oh hell, go read the rest of "Here's to the crazy ones" from Apple's Think Different ad campaign for yourself. Whether we're talking about Apple's target audience or the open source community that has had Microsoft running scared is immaterial. The point is that the DMCA, supported by this court ruling, prevents that sort of person from doing anything that's not sanctioned.
Report as a Journalist -- A third defense that Eric Corley's lawyers offered was that posting DeCSS was protected by the First Amendment's protection of the press, and by the First Amendment in general. It took the judge significantly longer to dispose of this defense, since free speech issues are notoriously tricky, but in the end, he concluded that the speech in this case is content-neutral due to the functional nature of the DeCSS code. He then went on to note that regulation of content-neutral speech is acceptable if it "advances the government's interests" and that preventing the copying of digital works is a government interest due to the existence of the Copyright Clause in the U.S. Constitution and the importance to the U.S. economy of exporting copyrighted materials.
If you haven't looked at the Constitution recently, the Copyright Clause reads, "To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries." Personally, I come down on the side of copyright existing to benefit society through the progress of science and the useful arts, and only secondarily to give authors and inventors exclusive rights. By my reading, the government interest thus lies in promoting the progress of science and the useful arts, and there's no question that the DMCA eliminates progress.
But I digress. The final result of the case was that Eric Corley and 2600 may not post DeCSS on their Web site or knowingly link their Web site to any other site on which DeCSS is posted. The decision was worded carefully so that linking in general would not be affected by the DMCA, but only in cases where "those responsible for the link (a) know at the relevant time that the offending material is on the linked-to site, (b) know that it is circumvention technology that may not lawfully be offered, and (c) create or maintain the link for the purpose of disseminating that technology."
In other words, it's acceptable to link to DeCSS if your intent is not to disseminate DeCSS, but merely to report on its availability, a fact I proved to my satisfaction with a trivial Google search on "download DeCSS" that provided over 17,000 hits, many of them still functional. You can verify this for yourself; just remember that DeCSS is only for Windows.
Here's where Professor Gillespie's argument becomes a bit more speculative. Although the court went no further in this case, he suggested that in any future cases in which the legitimacy of linking was called into question, he felt that the court would include in its deliberation the nature of the publication in question. For example, if the New York Times chose to link to DeCSS or some other technology that violated the DMCA (as in fact the San Jose Mercury News and Wired News have, in making the point that a ban on linking is seriously problematic), he felt that the court would have little trouble accepting the journalistic intent of the link. On the other hand, if some silly little electronic newsletter aimed at Macintosh and Internet users were to perform the same action, he was concerned that it would be more difficult to make the same defense. And if TidBITS wouldn't match up to the journalistic level of the New York Times in the eyes of a theoretical court, what about a blogger?
The end result would be that this court's interpretation of the DMCA could have the same effect of stabilizing the large news organizations in favor of the small newsletters and bloggers who are redefining what journalism means in today's Internet-enabled world. Speaking as someone who has done some of that redefining over the last 12 years, that worries me.
"Regime of Arrangement" -- In the end, Professor Gillespie argues that the true power of the DMCA is not so much related to its effect on copyright but these ways it weaves established organizations like large manufacturing corporations, research universities, and media conglomerates into what Professor Gillespie calls a "regime of arrangement."
Don't assume that these established institutions are necessarily being co-opted against their will. Apple's Think Different campaign reads like a manifesto for the very people who are disenfranchised under this regime of arrangement, and yet Apple is a member of the DVD CCA, and, obviously, a licensee of CSS for the DVD hardware and software that comes with the Mac. The open source community has proved the power of teams of independent programmers as an alternative to the traditional software development model, not to mention the ivory towers of research institutions. Distance education hints at the decline of the traditional university, and entrenched media organizations have struggled for years with the way the Internet lets anyone be a publisher.
If there's one theme we take into the 21st century, it's decentralization, and you can see it everywhere. The PC overtaking the mainframe, Napster changing the face of music distribution despite the recording industry's best efforts, DeCSS causing the movie studios conniptions, Linux successfully challenging the mighty Microsoft's server operating systems, even the terrorist attacks on the World Trade Center and the Pentagon - all are examples of the power of decentralization and the ever-increasing clash between these forces of decentralization and the centralized power structures that control everything about our world. I have no answers here, but I'd note that despite the awesome power of both systems, I'm seeing the forces of decentralization making significant inroads.
What Can We Do? I've been attending a number of talks on copyright and intellectual property issues at Cornell over the last year. Almost without exception, the talks are warnings of dark times ahead (obviously, most are slanted toward the academic and library worlds), but at the same time, none have offered any suggestions for how we can work to reverse the efforts on the part of the Content Cartel to lock up our cultural heritage and stifle innovation for the future.
At a recent talk by Alan Davidson of the Center for Democracy and Technology (CDT), I chatted with Alan afterwards about this problem, and he agreed it was a concern, but had no silver bullet to prevent the hordes of well-funded Content Cartel lobbyists from having their way with our elected representatives. I, too, have trouble knowing what will be effective, but I offer these possibilities.
I have to admit, I'm worried that none of this will be enough. The Content Cartel has the aura of celebrity on their side - they're "protecting" the rock stars and movie stars who sit at the pinnacle of today's society. They're the cool kids, whereas the people who campaign for civil liberties are often considered dull and overly earnest. My main ray of hope is that the reason most of the software industry voluntarily gave up copy protection technologies - primarily that consumers hated copy protection - will rise again, but unless we speak out now, all of our content may be locked up in a trusted system protected by the DMCA.
PayBITS: Is this is an important article on an important topic?
Non-profit, non-commercial publications and Web sites may reprint or link to articles if full credit is given. Others please contact us. We do not guarantee accuracy of articles. Caveat lector. Publication, product, and company names may be registered trademarks of their companies. TidBITS ISSN 1090-7017.