April 21, 2001

8:03 PM

You too can be a felon! Last year, the SDMI Foundation made a public challenge to see if anyone could crack 6 proposed protection mechanisms for digitally-encoded music. All six turned out to be feeble and all six fell. Since then, the SDMI Foundation has been relying on lawyers to cover up for the incompetence of their engineers. They're trying to suppress this article, so everyone reading this has a duty to make and store a copy of it. (Everyone should also own at least one copy of DeCSS. I have the 442-character C version printed on the back of my personal card.)
posted by Steven Den Beste (15 comments total)

By the way, "Kerckhoff's criterion" means that "security resides solely in the key". Auguste Kerckhoff was a military cryptanalyst in the 19th century, and his point was that no matter how obscure your cipher or code is, the algorithm will eventually be discovered by your enemy. If your system is strong, your communications will remain secure if the key remains secret. A good example of this is the RSA cipher, where the algorithm itself has been published (and revealed in a now-expired patent) but which is still considered practically unbreakable if a sufficiently large key is used. (Indeed, the beauty of asymmetric ciphers is that they remain secure not only if the algorithm is revealed, but also if the encryption key is revealed.)
posted by Steven Den Beste at 8:08 PM on April 21, 2001

What's with the asinine attempts to keep DeCSS public? Does it do anyone any good to make a song out of it, put it on a business card, t-shirt, tattoo, etc? If you don't have a working mirror up you shouldn't take so much pride in sticking it to the man.

I can almost picture future acheologists pondering what was so funny about 'all your base' and a significant amount of geeks putting some code in stupid places. It was cute at first, but we're loooong past that now.
posted by skallas at 8:39 PM on April 21, 2001

Skallas, it isn't about sticking to the man. It is very much about upholding your rights to free speech as far as I'm concerned. I pretty sure that the cuteness of the 1st amendment won't wear off anytime soon...
posted by fooljay at 9:19 PM on April 21, 2001

By the way, skellas, why all the harshness? "Asinine...you shouldn't take so much pride...geeks...stupid..." Why is it so important to you that you feel the need ridicule?
posted by fooljay at 9:25 PM on April 21, 2001

There's a serious point to it all. RIAA is trying to contend that "code is not speech" because code is for communication to computers, but speech is for communication between people. As such, RIAA is trying to claim that code is not protected by the First Amendment, which means that they can suppress code without violating constitutional freedom.

The people putting DeCSS on T-shirts, and writing HaiKu's, and songs, and all those other things are attempting to point out that this is incorrect by trying to smear the gray area between code and speech. For instance, if DeCSS is written in a unique synthetic language for which no compiler exists, then how can it be human-to-computer communication? If the algorithm is written in HaiKu, then it's clearly both poetry (art) and satire, both of which are clearly human expression. Moreover, a description of an algorithm (which RIAA also would love to suppress) is clearly "speech" since it's intended by one human to communicate information to another human. The principle that the anti-RIAA forces are trying to establish is that speech is communication, that communication involves moving information from one person to another, and that the encoding of that information doesn't matter -- thus it is just as much speech to write an algorithm in C as it is to write it in Jive or Hai Ku or structured English.

DeCSS itself isn't important. But the underlying principle of the First Amendment is overwhelmingly important. This is part of the age-old fight against those who, for whatever reason, would like to censor us and to control our thoughts. We must always be on our guard to prevent the forces of censorship from chipping away our freedom of expression.
posted by Steven Den Beste at 9:26 PM on April 21, 2001

I don't see the validity of taking DeCSS code and altering it in a way that's no longer C++ (or whatever its original form was) for some legal fine-lining. Its already been declared illegal which I think the appropriate response would be to put it up everywhere (working mirrors) to show protest and provide accessibility to software DVD player developers, not stamping it on a jellybean to somehow equate code with a convulted free speech argument.

Certainly I'm against cersorship of this kind - this kind being corporate interests before human rights, but at the same time we can go backwards from the 'code is speech' position to repeal most if not all copyrights on software.

I also don't see the RIAA's troubles are with speech laws as much as those troubles are with the defendants. The RIAA is using existing intellectual property laws (as screwed up as they are) to keep DeCSS from the public. Right now I see two kinds of protest, intelligent ones (mirrors, awareness raising) and silly games (Rot-13 DeCSS on the goodyear blimp,etc.)
posted by skallas at 10:51 PM on April 21, 2001

Also, your post asks us to make a copy of the SDMI article. What if my copy was on a toothpick, how useful would that be? Compare that to a web mirror.

In the IP vs. free speech/free access argument toothpicks, cards, and their ilk may be cute but don't get anything done.
posted by skallas at 11:02 PM on April 21, 2001

Its already been declared illegal which I think the appropriate response would be to put it up everywhere (working mirrors) to show protest and provide accessibility to software DVD player developers, not stamping it on a jellybean to somehow equate code with a convulted free speech argument.

It's not a convoluted, just a reductio ad absurdum attack on the MPAA's arguments. Fighting the free speech case here is important, simply because the MPAA is trying to suppress constitutionally protected activities. This is opens the door to suppressing other speech - if the MPAA wins, how long would a detailed set of instructions for breaking CSS remain legal?

It's also important to legally establish that there are very strict rules to declaring speech illegal. In the case of DeCSS, the code may or may not be used for legal purposes and, during the first case, the MPAA couldn't point to a single example of piracy because of it. Setting the precedent that speech could be suppressed because of possible misuse would open a lot of nasty possibilities.

we can go backwards from the 'code is speech' position to repeal most if not all copyrights on software

How so? The only way that could be true is if software had some special copyright status that didn't apply to the other forms of speech. Copying a book is illegal even if it's protected speech.
posted by adamsc at 2:19 AM on April 22, 2001

I think the explanation lies in the way the USA works - there is this sacred constitution that acts as the "nuclear option" in any debate. So there's a strong incentive for trying to tie whatever you are doing to something in there - hence code + free speech + t-shirts.

That and people wanting to demonstrate their allegiance to a particular group, I guess.

I find all these arguments a little confusing - why so much emphasis on DeCSS rather than gene-related technologies, for example? DeCSS seems to be of relatively little importance compared to the right to use our own genetic code (the code/speech thing was more than adequately made by this book - but of course that doesn't help people avoid paying for movies etc).
posted by andrew cooke at 3:21 AM on April 22, 2001

Copyright is not involved here. If there were source which was written by RIAA, then it would be covered under copyright. If someone stole that source and published it, that copyright would be violated and the court case would be very straightforward.

What RIAA is trying to do is much different. Someone independently developed the concepts behind DeCSS and wrote the code themselves and distributed it. RIAA is trying to suppress something which someone else owns the copyright on.

Or in the case of the SDMI crack, RIAA is trying to suppress an academic paper which describes how certain people approached and cracked the SDMI watermarks. This has nothing to do with copyright, but everything to do with thought control.
posted by Steven Den Beste at 5:43 AM on April 22, 2001

Granted, there's certainly not much practical value in having DeCSS on a shirt or keychain or toothpick or whatever... but I still love the idea! To my mind, it's a really elegant way of saying "once information is created, it is fluid and free; one cannot easily destroy or hide it." It brings me a certain feeling of power and satisfaction in the face of stupid selfish greediness.

So, yeah, Skallas, you're basically right that there are two approaches happening here: practical ones and "silly games"-- but (at least) for me, silly games are very important.

Dig?

=t=
posted by JimmyTones at 7:59 AM on April 22, 2001

The confusion here is over the fact that there is more than one argument going on, more than one right at stake. The right to use and post DeCSS is one issue--namely, fair use--and it's what skallas says we should focus on. The people putting DeCSS on T-shirts, in poetry, embedded in DNS entries, or whatever are arguing about a different, but related point: that code is speech and can't be suppressed.

The DMCA violates our right to fair use of digital media, but it's doing it through supression of Free Speech. So if you're going to win a case about DeCSS, which is the more powerful argument? Since Free Speech is directly embedded in our Constitution, I would say that's the better argument to make. The people playing "silly games" are fighting the Free Speech aspect, because the concept of fair use has pretty much been written out of law by the DMCA.

why so much emphasis on DeCSS rather than gene-related technologies, for example?

DeCSS is a case that's being tried right now. Are there any comparable cases regarding ownership of genetic information? DeCSS is in court now, and that makes it a prime target. It's also a lot easier to understand the implications of.
posted by daveadams at 10:34 AM on April 22, 2001

Steven, where can I find that 422 charachter version? I'm designing my new business cards this week, and I feel that it would be fun to put on them.
posted by SpecialK at 10:47 AM on April 22, 2001

There used to be a page at Cryptome which contained as many gray representations of the CSS algorithm as possible, but I can't find it now. I'd post it here except I won't do that without Matt's permission. (In the mean time, I'm mailing it to you.)
posted by Steven Den Beste at 12:45 PM on April 22, 2001

Ah. Here's that short C program. Here's the whole gallery I thought was on Cryptome but which was actually somewhere else.
posted by Steven Den Beste at 2:22 PM on April 22, 2001