Slashdot Mirror

Yesterday in a toasty courtroom in lower Manhattan, Stanford Law School dean Kathleen Sullivan faced off against lawyers for the world's biggest movie companies and a lawyer for the U.S. Justice Department with oral arguments in the appeal of the 2600 case. One of the three judges hearing the case -- Jon Newman -- appeared to be the designated questioner. He asked nearly all of the questions in both this case and the ones heard earlier in the day. He probed both sides about equally, trying to find flaws in the arguments of whoever was speaking at the time. I'll cover the hearing below, and there's possibly a few areas where the Slashdot crowd could assist in the case.

Sullivan spoke first. She argued that since the DMCA restricts speech, under the First Amendment the government must narrowly tailor the law to only restrict those specific areas of speech that it is targeting. Furthermore, the government bears the burden of proving that the speech it is restricting is a problem in some way -- usually it does this by holding hearings, getting testimony, etc., in the process of passing a law. She noted that none of this was done for the DMCA, and that the DMCA restricts many areas of speech that cannot constitutionally be restricted.

She also made much of a rather telling fact: there is no piracy attributable to DeCSS whatsoever. Not one traditional copyright infringement has ever been attributed to DeCSS, and the movie studios admitted in the case that they could not produce even one example of an infringement due to DeCSS. (Technically-literate people may realize that mass DVD copying is performed by stamping complete copies of the DVDs, encryption and all, no decryption required, though that wasn't covered in the hearing.) But Sullivan jumped on this point for all it was worth and then some -- the judges seemed fairly skeptical about accepting it, trying to insist that widepsread and massive copyright infringement due to DeCSS must be occurring, somehow, somewhere. It just must be.

She ran into her first really hard question when she stated that computer programs were expressive, and the judge asked her to explain. Her answer was that programs were beautiful in and of themselves, that they could represent scientific research, that they could be poems, and that they could do things -- their functional nature. I felt the response was lacking. Sullivan managed to work in the recent ruckus over a Princet on scientist unable present his work due to DMCA threats, which was cunning of her. If a Slashdot reader can create a pithy and short explanation for how and why a computer program is expressive speech and/or what it expresses, it might be useful.

Sullivan also argued that under free-speech precedent, if less restrictive alternatives were available to the government and it failed to use them, the law must be overturned. The judge mentioned the Audio Home Recording Act -- the law passed in 1992 which both implemented serial-copy protection in digital audio tapes and explicitly legalized home taping. Sullivan pointed to AHRA's serial copy prevention as an example of a law which restricted copying but which was not as restrictive as the DMCA turned out to be. This argument seemed to be pretty powerful with the judge.

The next point to be discussed concerned the injunction issued by Judge Kaplan, and his written opinion in the case. The Appeals judge made the point that the injunction could not be considered to apply to anyone except the specific defendants -- that is, just because 2600 was enjoined from posting or linking to DeCSS, doesn't mean that anyone else necessarily would. On the other hand, the reasoning applied in the opinion could be assumed to apply to other U.S. citizens wanting to post DeCSS. The gist was that Sullivan couldn't argue her case as if anyone would be enjoined from linking to DeCSS, but only regarding the specific defendants that were.

Finally they got to the idea of "disseminat ion," since the DMCA prohibits dissemination of circumvention devices. What does disseminate mean on the internet? The judge and Sullivan agreed that the New York Times is in the business of disseminating information (the NYT being today's quintessential example of "the press"). The judge asked if the New York Times intends to disseminate all of the information on every page it links to in its online edition. Sullivan said yes. The judge asked if the NYT specifically intends to disseminate every bit of info on every single page that it ever links to -- again Sullivan said yes.

Assistant U.S. Attorney Daniel Alter was up next. He started with a hypothetical: What if someone developed a program that could shut off the navigation system in commercial airplanes? What if someone developed a program that could shut off smoke detectors in public buildings? Surely, he said, the government could ban the publication of programs which were a threat to people's lives. He proceeded with the standard quotable rhetoric: DeCSS is a "digital crowbar." Hey, if you're a reporter covering the case and you don't understand it, at least you got a phrase that jumped out at you screaming to be quoted.

He then got down to the meat of his argument -- that the government can regulate conduct even if there's a speech component to it. He used the example of Giboney V. Empire Storage and Ice Co., a case where picketers (a constitutionally protected activity) were successfully prevented from picketing due to the functional intent of the picketing, which was apparently to violate certain laws relating to restraint of trade. Alter argued that the DeCSS case was similar -- the intent of distributing DeCSS is to promote violations of copyright law, therefore the speech part of such distribution can be ignored by the courts and the courts can focus on regulating actions without concerning themselves about speech issues.

Alter proceeded to postulate that the government has the ability to create and regulate a market in expression, and correct any market flaws that may exist. Viewed from this vantage point, the existence of the Internet and all of those unrestricted personal computers connected to it is one large market flaw which the government has the power to correct. He used the example of must-carry laws for cable systems -- cable television must carry local broadcast channels, and the official reasoning behind that is that otherwise cable systems would drive broadcast television into bankruptcy and the government is preserving a vibrant market in broadcast television through the must-carry laws.

He stated flatly that the problem with digital works is that they can be copied. He argued that the DMCA is actually pro-First Amendment, as a means to promote the market for digital works. So in the calculus of the government attorney, increasing the speech of a dozen movie studios at the cost of decreasing the speech of 260 million citizens is a win for the First Amendment.

The judge asked about the Audio Home Recording Act and serial copying -- why wouldn't the "no serial copies" approach taken to DAT recordings with SCMS under that law represent a less restrictive means for the government to promote copyright in the digital age? The attorney argued, of course, that the DAT law was inapplicable since it predated the massive growth of the Internet -- and this is where he pulled a fast one on the court. Alter stated that, due to the Internet, one only needs a single copy for "catastrophic" infringement, so even that one copy permitted by the Digital Audio Tape serial copy scheme would be too much. One copy, the judge asked? Yes, he said, just one copy and put it on the Internet and ...disaster. Apparently, in the attorney's world, once that lone copy is made, it pretty much automatically puts itself on the Internet with no further acts by any individual. The point Alter narrowly evaded evaded it is that the act of publishing a copyrighted work to the world is a copyright violation in the traditional sense, and is punishable under traditional laws.

So, the judge said, Congress needs a more restrictive technique to prevent copyright infringement because the Internet is now a factor? The DA claimed that it does.

The judge next moved to one of the most interesting questions of the day -- does fair use require access to a work in its original form? That is, one cannot excerpt a digital clip of a CSS-encrypted DVD, but one could point a video camera at the screen and create a clip, albeit of poor quality. Is that sufficient for fair use? This question has disturbing ramifications, depending on who is asking it and how it is answered. It seems odd, at first glance, to insist that one must be able to make fair use of a work in its full, unfettered, most-advanced, highest-quality form. But after thinking about it for a bit, I realized that anything else utterly destroys fair use. What if I could make clips of 256 kilobits/second mp3s, but the clips were at 16 kilobits/second? Would that be sufficient? Is a 16 kilobit/second mp3 even recognizable as music? What if book publishers could designate the Swahili version of a book as the "fair use" version, and completely shut down any quoting from the English version -- ("After all, you can still quote freely from the Swahili version; it may have a few words missing, and it's in Swahili of course, but you can still quote from it.") The judges seemed to be actually considering that filming a DVD movie from the television set or getting some macrovision-corrupted analog output might be sufficient for fair use purposes, and I hope they think it through and reject that idea entirely.

The attorney moved on to linking. He argued that 2600's actions ought to be examined in their entirety; that 2600 was effectively "shuttling" people over to commit a crime by linking to the DeCSS code. According to him, the entire conduct of the defendants should be considered to divine the purpose behind linking to the DeCSS code. If it were for some legitimate purpose, a link would be okay. But if the purpose were to "shuttle" people to commit a crime, that wouldn't be. The number of links would be important, the context would be important, and the intent of the writer would be important to this analysis. Search engines, according to the attorney, would be okay they are just providing lots of links without the harmful intent that the attorney felt was necessary. So apparently something like this:

"This is a scholarly discussion of DeCSS. We are a major media outlet, and would never encourage lawlessness, so this link to DeCSS is okay."

... is fine, while this:

"Hey all you l337 h4x0rz, come get DeCSS and use it to copy movies and watch them automatically distribute themselves via the Internet!"

... is not. How context works, I'm not sure. Certainly the vast majority of 2600's links that it has ever published are not "shuttling" people to copyright infringement -- the vast majority are for the standard journalistic purposes of disseminating information. But somehow under Alter's analysis, 2600 came up lacking while the NYT did not.

The judge cut deep with a hard question: "Can you prosecute a newspaper who publishes a list of stores where obscenity can be purchased?" The parallels to this case should be obvious. The attorney dodged the question with an outstanding answer: "Yes and no." He tried to go back to his theory of looking at the overall conduct of the newspaper, but it was clear that he didn't want to say "Yes, we can prosecute the newspaper for publishing the list of stores" but did want 2600's actions to be covered, and wasn't sure how to reconcile those two desires ... and neither were the judges. I'm not sure they bought his argument.

Finally, Charles Sims, the lawyer for the MPAA.

He had had time to pay attention to the previous efforts and tailor his argument somewhat. He tried to cover weak areas -- insisting, for instance, that no record of harm is required for Congress to regulate pure speech. He brought up the Congressional record (hearings, testimony, etc.) that pre-dated the DMCA, and said it showed "actual harm" to the movie industry.

Actual harm, the judge asked? "Yes. Actual harm," he replied. "Well, actual threat of harm." That got a laugh from the audience, and scored him no points with the judges. He didn't use the "digital crowbar" metaphor, but insisted that publishing DeCSS was like publishing the combination to a bank vault in a newspaper -- something which is not, as far as I know, a violation of any law, though it might well inconvenience the bank.

The judge asked this lawyer too the hard question about less restrictive means to accomplish the same goal and serial copy management. The MPAA's tactic was similar but slightly different than the U.S. Attorney's; the AHRA is inapplicable, he said, because Congress didn't take the Internet into consideration when drafting it. He also argued something that will make him no friends with the RIAA -- that motion pictures deserved more and better protection than music (so the AHRA serial copying wasn't appropriate for movies). After all, he said, motion pictures have never been subject to the sort of fair uses that music has, the copying and so forth. I suppose he doesn't own a VCR. This argument about motion pictures being more deserving than music seemed strangely surreal -- for the first several decades of motion pictures, they had much, much weaker First Amendment protection than other forms of speech because the courts considered them to be solely entertainment, and only an assortment of free-speech challenges to laws restricting them earned them the privilege to stand on a par with other forms of speech in the protection of the First Amendment. Now, the motion picture people are not only arguing that their form of speech is more privileged than others, but they're arguing that still another form of speech, computer programs, ought to be considered in that inferior, functional category that motion pictures worked so hard to escape from. It's a strange world we live in.

The judge asked whether the DMCA created a "permanent" copyright, or an effective extension of copyright. The lawyer smoothly dodged the questions by saying that movie studios could (not "would," but "could") publish works in unencrypted form when (if) their copyright on the work ever expires, or perhaps someone could use a decryption device then, since it would no longer be illegal under the DMCA to do so. The judge asked where those encryption devices would be, after all, they've been banned by the DMCA. The lawyer had faith that they would appear. So apparently: the fact that the studios haven't gotten encrypted content working in an impenetrable fashion yet means that they aren't screwing you out of your access to works when copyright expires.

In closing, the MPAA lawyer compared CSS to one putting a painting in one's living room or charging admission to a movie theater to see a movie. But the right to exclude people from your living room or a movie isn't created by copyright law, it's created by property law -- your home is your home, and you can exclude people from it to your heart's content. The MPAA's conception of property law was that the movies they release are essentially their home, and they have an absolute right to do anything they want with this property until copyright expires. It is a nice sleight of hand to conflate one's right to one's home, perhaps one of the most powerful rights a citizen has, with one's right to control how a movie is viewed is someone else's home. He seemed to be hoping that the one would rub off on the other.

In closing, Sullivan had a brief rebuttal period. Not worth going into; she tried to call the other two lawyers where she thought they went too far astray and she could zap them.

The judges took the case. They also requested one last brief from both sides, due by May 10th, to cover anything that came up at the hearing and the parties think needs to be explained further. I would suggest that it's likely that the people who draft the brief will read this article; and that insightful comments could be of assistance. I think there are a couple of key areas which people may be able to answer:

1. Why and how is a computer program expressive speech? What does it express? 2600's lawyers are entirely familiar with Touretzky's Gallery, so forget about those. Assume you have some C or perl staring at you, any random block of code in any random print-out. What does it express? Why should that code be protected expression?

2. What examples of fair uses absolutely require access to the work in its most modern, digital, uncorrupted, un-macrovisioned form? The only one that jumped out at me is making a backup copy in case the original is destroyed. But perhaps there are others.

Reader Trinition also points to this brief a ZDNews article on the hearing; the case was well-attended by the press and by people like the members of LXNY, New York's Free-software organization, so there are quite a few personal and press accounts around the Net.

twrayinma writes: "According to this article in Salon, Prof. Felten may have been smarter than we thought yesterday when he "allowed himself to be bullied" by the DMCA wielding RIAA. 2600 and the EFF could use this incident as an example of a big company using DMCA to quash legitimate research, as their court case alleges. Looks like Prof. Felten's fight may not be over yet..." Carl Kaplan's Cyber Law column in the NY Times also discusses the case.

While I'm at it, I should mention that the appeal hearing in the 2600 case will be this Tuesday in New York City. I do not know if a "protest" of any sort is planned; judges aren't supposed to decide cases based on protests anyway. But the case may be of interest to many NY-area readers:

Tuesday May 1 2001
Second Circuit Courthouse
10:00 AM
Courtroom 506
40 Centre St., at Foley Square, Manhattan

Thanks to LXNY for the information.

gte024h writes: "I have been approached by a friend working at a mid-sized corporation about a plan to salvage hardware from their "dead" computers to make working computers which would then be donated to a local school. In the past I have salvaged old hardware from them, but only a few machines at a time. With this plan I will initially get about 40 identical machines with perhaps many more later (he says they have a warehouse full). I have no idea what condition these machines will be in but I am pretty sure that out of 40 non-functional machines I should be able to get 10-15 working. Does anyone have experience doing anything like this? Would it be faster to strip the machines and test the components individually, or to troubleshoot each machine whole? Any ideas for efficient testing and such are greatly appreciated."

I know that volunteers from LXNY recently did something similar with a whole boatload of donated computers; anyone with experience care to comment on how to best handle the mixed blessing of donated, but old, hardware?