FITUG e.V. Förderverein Informationstechnik und Gesellschaft

[GILC-plan] Article: Copy Controls and Circumvention: Do

------- Forwarded message follows ------- Date sent: Thu, 17 Jan 2002 15:00:35 -0500 (EST) From: Andy Oram <andyo@oreilly.com> To: gilc-plan@gilc.org Subject: [GILC-plan] Article: Copy Controls and Circumvention: Don't Get Around Much Any More

[ Double-click this line for list subscription options ]

http://www.oreillynet.com/cs/weblog/view/wlg/1068

Copy Controls and Circumvention: Don't Get Around Much Any More

by Andy Oram Jan. 17, 2002

The one-two punch delivered by two court rulings in a closely followed DeCSS case failed to knock out [50]2600 Magazine and publisher Eric Corley, despite predictions from many observers that the final round was over. Instead, on January 14, the defendants rallied and petitioned for a rehearing, with legal support as always from the Electronic Frontier Foundation. Their petition, superficially a series of hair-splitting definitions and references to court rulings, actually contains a powerful argument that may be used to rule unconstitutional the notorious anti-circumvention clauses of the Digital Millennium Copyright Act.

The 2600 case was brought by movie studios to prevent the spread of DeCSS, which would permit unauthorized access to films on DVD. But the case affects the development of innovative technology, the publication of programs and computer science information, and the Web's fluid nature reflected in the freedom to make links.

In this article, as a non-lawyer interested in the development of new media, I will try to explain the complex reasoning behind the Second Circuit Court of Appeal's [51]ruling against 2600 and the defendants' [52]petition, officially an "application for rehearing en banc."

All you really need is heart

The case brought against 2600 was just one of several instances where speech or information exchange has been constrained by the "anti-circumvention" clauses of copyright law. The most extreme case was the arrest of Dmitri Sklyarov, who had come to the United States from Russia to discuss his research at a computing conference. He was jailed for several weeks in a case brought against his company for a product that circumvented access controls on documents in the Adobe Acrobat eBook Reader format. Another high-profile case is that of teenager Jon Johansen, indicted in Norway for whatever role he played in developing or disseminating DeCSS.

Passed by Congress in the DMCA in 1998 and listed in the U.S. Code as Title 17, section 1201, the anti-circumvention clauses talk of illegal "devices" and of "access to a work protected under this title." The thrust of the law appears irrelevant to issues of free speech: it seems to address the copyright equivalents of devices that warn drivers about police radar or the old, illegal "blue boxes" that permitted people to make free long-distance phone calls.

Yet Congress was well aware that copyright functions in certain circumstances as a restraint on speech. The legislators inserted clauses in the DMCA explicitly claiming that it should not affect free speech or fair use.

I am about to delve into fussy details of histories of interpretations of fine points of copyright law. One might be tempted to throw all these minutiae overboard and judge the issue from a naive or common-sense standpoint. If you do, you're in good company. A leading lawyer's guide to copyright for lay people, Jessica Litman's [53]Digital Copyright, suggests bringing copyright law in line with the public's common-sense understanding of the proper use of a work. But the issues in the DeCSS cases are too tangled to yield the truth to that high-minded approach.

I think it's pretty clear that DeCSS is the kind of thing Congress intended to prevent when it passed the DMCA. DeCSS defendants cases tend to argue that the law has the unintended effects of suppressing speech, competition, and other things to which the public has rights. Congress never said it wanted to do these things (in fact, the law suggests the opposite). However, only one member of Congress--Representative Rick Boucher, a frequent champion of open Internet development--has questioned them, either. No one else has decried the chilling of academic research that Edward Felton claimed to have suffered, or the bans on Web links, or the arrest of Sklyarov (which legitimately could have turned into an international diplomatic incident). Furthermore, this kind of outcome had been predicted by critics of the law before its passage.

Therefore, despite fig leaves to cover fair use and other rights, Congress is obviously content with the outcome of the law. So is the executive branch of government, if one judges from its harsh prosecution of Sklyarov and from the Copyright Office's narrow rulemaking last year on exemptions from the anti-circumvention clause.

Congressional intent does not make something right. Indeed, many critics are trying to establish that Congress overstepped its constitutional power by imposing such a broad ban. That is why legal arguments matter. In particular, the Second Circuit ruling has already been cited in another case in which the plaintiff seeks to prevent the distribution of DeCSS code as a violation of trade secrets.

Today is the last day that I'm using words, don't function anymore

The main barrier the court had to leap in order to rule against 2600 was the argument that computer programs were speech and that their distribution was therefore protected under the First Amendment. This doctrine was used in two high-profile cases dealing with a different kind of program, cryptography. The doctrine was affirmed in reference to DeCSS code by a November 1 ruling in California (but in a different case from the 2600 case considered here).

While laying out and affirming the reasoning in these previous cases, the Second Circuit permitted itself the luxury of overriding free speech issues through a series of assertions, all of which are based on previous cases and appear in previous rulings on this case:

1. Computer code is not just speech, where it serves an expressive purpose; it also serves a functional purpose.

2. The anti-circumvention clauses ban the functional aspect of DeCSS, not the expressive part. This makes the ban "content-neutral."

3. Therefore, the courts have a relatively wide leeway to ban DeCSS. They need not be as careful to protect the First Amendment as they do when banning the expressive aspects of speech.

A distinction between the "expressive" and "functional" aspects of speech winds its way through numerous court cases, including the ones on cryptography mentioned earlier. But computer code is not the only kind of speech that is functional; recipes, directions from an appliance manufacturer, and many other forms of speech also have functional aspects. Consider shouting "Jump!" to a person on a window ledge or "Hang him!" to a mob who has captured a criminal.

Since the "content-neutral" assertion was so critical to the ruling, it deserves a bit of history and context.

A classic case of a content-neutral restriction on speech is a ban on the use of megaphones within an urban area. The content of the speech is irrelevant; the goal is to prevent public nuisances. If the city allowed the Girl Scouts to use megaphones but prohibited the Nazis from doing so, the ban would not be content-neutral and would therefore be unconstitutional.

By this logic, there could not possibly seem to be any argument that a ban on DeCSS is content-neutral. Downloading and running Internet Explorer on your computer is legal; downloading and running DeCSS is illegal. That's a fundamentally content-based ban.

The judges circumvented this First Amendment issue by saying that they were banning DeCSS for its functional aspect. They were not banning it because the code contained obscene words or promoted violence; they were banning it only because it decodes movies. That made the ban content-neutral.

Those of us who are not lawyers may feel as if we just passed through a lesion in the space-time continuum. Code is speech and (according to some courts) code is functional, but it is exactly the same code. It is like box that is longer than it is wide. If the long (expressive) side does not fit through a doorway, the court is free to turn the box on its narrow (functional) side to shove it through. Professor Lawrence Lessig writes, "The test is to discover what the real purpose of regulation is. If the purpose is to silence certain ideas, it's content-based. The anti-circumvention clause is not meant to silence ideas and therefore is not content-based. That's the thrust of the court's opinion, and it draws into question the court's general approach, but perhaps not the particular application."

Wielding the functional and content-neutral assertions, the Second Circuit rooted its argument in the plaintiffs' camp. The judges thereby reaped an extra bonus. They still have to deal with First Amendment issues, because they agree that computer code is expressive as well as functional. But they don't have to apply "strict scrutiny" and find the "least restrictive means" to achieve their goals, as they do in typical First Amendment cases involving erotica or other expressive material. Instead, they merely need to apply "intermediate scrutiny." (I am indebted to Professor Peter D. Junger, of Junger v. Daley fame, for this point.)

That gives the court much more leeway to impose a ban. Enough leeway, in fact, to permit the meat-cleaver injunction banning Web links to sites containing the offending code. This part of the 2600 ruling (which has been in effect for some time, and obviously is not content-neutral) is the strongest precedent yet in the alarming creep of legal doctrine toward restricting Web links, or making sites liable for the material on the sites to which they link. A link is now "functional," and therefore subject to more restrictions than normal speech. (The defendants' appeal points out that the courts here are going far beyond the DMCA, "which nowhere in its text or legislative history refers to hyperlinks.")

Any computer code, unless it consists of an empty pair of braces, is functional. Therefore, if the Second Circuit's ruling holds, all computer programs will permanently be given less protection than other forms of speech. This almost certainly was not the intent of the courts that previously declared computer programs a form of speech, and the Second Circuit did not openly repudiate those rulings.

The defendants' appeal argues that the Second Circuit made the wrong choice, even if the standard of "intermediate scrutiny" is accepted, for several reasons:

1. First (to attempt an explanation in everyday language), the court has to figure out what harm it is trying to prevent. But there is no harm; no one has used DeCSS to illegally copy a DVD. DVDs are easy to copy in other ways (or least, so claims the unsolicited email that I have been receiving daily for several weeks). But as far as DeCSS goes, in the words of the appeal,"there was no demonstration of actual harm." This is why, elsewhere, the appeal says: "the panel incorrectly failed to scrutinize the empirical record for `substantial evidence' to support the burden on speech."

2. Second (and here is the powerful argument on constitutionality I mentioned earlier) an earlier case has ruled that the court is supposed to look for "the availability and efficacy of `constitutionally acceptable less restrictive means' of achieving the Government's asserted interests." There are plenty of things (itemized in the appeal) that Congress could have done to prevent copyright infringement that are less restrictive than the proven effects of the anti-circumvention clauses.

In short, the defendants are telling the courts to do their job and uphold the First Amendment in the face of a radical, sloppy, and repressive law. And the defendants have found precedents to back up their challenge.

There is good grounds for pessimism, though--the courts have let the large studios lead them around by the nose, just as Congress has. The trend toward finding any excuse suitable for suppressing the distribution of programs that could be used to exchange copyrighted material will soon have an even more substantial impact. The music studios had a strong case when they shut down Napster for vicarious and contributory copyright infringement. But now they are trying to shut down sites that distribute purer forms of peer-to-peer file-sharing programs.

The creators and distributors of the programs do none of the indexing and connection-making that Napster did. Their only involvement in copyright infringement is to distribute a program that has many useful and legal purposes. But the copyright juggernaut has become unstoppable. Watch the courts find a way to clamp down on the trade of computer programs and to further weaken any rights left to programmers, product developers, and the public.

Nice work if you can get it

Some defenders of DeCSS suggest changing copyright law so that the anti-circumvention clauses are applied only when actual copyright infringement takes place. This would make the anti-circumvention law less of a radical imposition on the course of technology. Perhaps it would change an unconstitutional law into a constitutional one. But it would leave the courts to decide the programmer's intent, something that is hard to determine even with DeCSS.

The DMCA contains two passages that provide fodder for its critics by suggesting that Congress did not intend the law to have the momentously damaging impact that is now emerging. The passages are striking because they were clearly inserted to reflect limits requested by opponents or by those worried about unintended consequences, not because they are integral to the law. The Second Circuit quickly disposes of these passages with an attitude that appears to be: "if it invalidates the law, it must be ignored."

One passage is:

Nothing in this section shall enlarge or diminish any rights of free speech or the press for activities using consumer electronics, telecommunications, or computing products.

One would think that this passage was meant to act as a brake upon abuses of the law, and indeed the defendants cite it in defense of their free speech rights. The Second Circuit brushes past the sentence with the excuse that, "Congress could not `diminish' constitutional rights of free speech even if it wished to." End of discussion, except for a bit of finger-shaking at the defendants for trying to "enlarge" their free speech rights.

Taken out of context, the court's argument has logic. Congress isn't abridging free speech because it can't do so--yet it allows the movie studios to abridge free speech through lawsuits. The computer code is less subject to First Amendment protection, as explained before, because of its "functional" aspect and the "content-neutral" ban.

A comparable passage is subsection 1201(c)(1):

Nothing in this section shall affect rights, remedies, limitations or defenses to copyright infringement, including fair use, under this title.

The court's reading of this clause is even more cynical than the previous one. It says that fair use applies to the copyrighted material itself, not to the circumvention of technical measures. And only circumvention, not use, is prohibited by the law. According to this court, once you've illegally circumvented the encryption system, according to this court, viewing the material is legal:

Subsection 1201(c)(1) ensures that the DMCA is not read to prohibit the "fair use" of information just because that information was obtained in a manner made illegal by the DMCA.

Thank heaven for small mercies.

To its credit, the Second Circuit made a consistent argument that the DMCA does not abridge fair use. To be sure, studios could prevent a history professor from showing a segment of a movie to a class, or prevent a radio critic from airing snippets of an audio program along with commentary. But the professor could capture a segment by stationing a video camera in front of a screen, while the radio critic could play the audio program into a microphone. The resulting degradation in quality, the court claimed, would probably not impair the educational or critical use of the work.

To be sure, the court approaches fair use more respectfully the entertainment industry, which openly considers fair use a fluke that is past its time. But one can detect a cavalier attitude toward the public in the court's suggestion that our legally protected rights be achieved through such cumbersome measures. One could expect a court dedicated to preserving legal rights and public interests to suggest a more dignified course, such as requiring licenses to include provisions for allowing users to replay portions of a work for critical and research purposes. Furthermore, its remedy does nothing for libraries and other institutions that need to archive a work.

But the degradation of visual or sound quality is overshadowed by a more subtle degradation in community and academic life. By making untraditional uses of works more difficult, the DMCA will hamper people who want to bring something to an informal meeting of their church or civic group. The casual information sharing that provides new ideas and impetus for students and community activists will markedly decrease.

How long do I have to wait?

The Second Circuit's opinion contains a good deal of disconcerting reasoning that I'll mention here without going into detail. For one thing, it carefully screens facts to present a history of the DMCA and the DeCSS case the way the studios would want it. In addition, it continues the tendency I've noted earlier to ignore what it can't fit into its scheme.

The defendants argued the DMCA unconstitutionally overrides copyright law, because technical controls last forever while copyright is meant to expire after a certain time period. (Of course, this is somewhat moot because the deadline keeps being extended by Congress as a favor to people who died decades ago.) The Second Circuit dismisses this self-evident criticism in a characteristically perfunctory manner: "the argument is entirely premature and speculative at this time on this record." Then the judges relent a bit by citing the earlier decision in the case:

As Judge Kaplan noted, the possibility that encryption would preclude access to public domain works "does not yet appear to be a problem, although it may emerge as one in the future."

So when can we challenge the law? Seventy years from now, when some work is effectively prevented from entering the public domain?

Even given the Second Circuit's predilection to rule for the plaintiffs, its insensitivity to the seriousness of its ruling is cause for concern. The ban on linking, for instance, is a major intervention into the rights and practices of the Web--in fact, a blow at its very heart--not to mention a shadow hanging over communications technologies that will emerge in the future. The new limitation of free speech in computer programs is also far-reaching.

As the defendants' appeal points out, the Second Circuit ruling relegated Internet activities to "second-class First Amendment citizenship." The famous Supreme Court ruling against the Communications Decency Act, Reno v. ACLU, pointed out the boon represented by the Internet's speed and ease of connection. But these are precisely the reasons that the Second Circuit considers links dangerous! Thus the appeal says that "the panel stood ACLU I on its head."

The court prided itself on its "evolutionary" approach, its "narrow" holdings, and its "appropriate caution." But this narrowness applies only to the rights of the defendant. The plaintiffs are running amok. If there's one thing we've all learned during the long fight over the DMCA, the WIPO copyright treaty, and the cybercrime bill, social effects of legislation tend to have long reaches.

[54]Andy Oram is an editor at O'Reilly & Associates specializing in books on Linux and programming. Most recently, he edited [55]Peer-to-Peer: Harnessing the Power of Disruptive Technologies. ... Copyright (c) 2000-2002 O'Reilly & Associates, Inc. All Rights Reserved.

References

50. http://www.2600.com/ 51. http://www.eff.org/Cases/MPAA_DVD_cases/20011128_ny_appeal_decision. html 52. http://www.eff.org/IP/Video/MPAA_DVD_cases/20020114_ny_2600_appeal.h tml 53. http://www.digital-copyright.com/ 54. http://www.oreillynet.com/pub/au/36 55. http://www.oreilly.com/catalog/peertopeer/ _______________________________________________ GILC-plan mailing list GILC-plan@mailman.gilc.org https://mailman.gilc.org/cgi-bin/mailman/listinfo/gilc-plan

------- End of forwarded message -------

Zurück