Just a Number

by on May 3, 2007 · 26 comments

I mostly agree with Tom Lee’s point here, but I think he’s being a little bit unfair in his characterization of Ed Felten’s post on the AACS/Digg incident. Tom says:

I’m no fan of DRM, and I think the AACS LA’s actions are pointless and stupid. But Doctorow and Felten are being disingenuous — they’re simply too smart not to see the problem with this argument. Namely, that any type of data, sampled at a chosen level of precision, can be represented as a number. Consequently, if you believe that one or more types of information deserve legal protection — as Felten seems to, when he refers to songs & movies — then the argument that “it’s just a number!” becomes ridiculous.

Sixteen bytes is probably too short to merit a copyright. But that’s not the right that the AACS LA is asserting: they’re calling the code a “circumvention device” under the DMCA. And even if you don’t recognize the DMCA’s validity, there are other forms of intellectual property protection that may apply — there are laws related to trade secrets, for example. If you just think about it a little, it should be obvious that even a very short piece of data can enjoy some kinds of legal protection. Sixteen bytes is more that enough room to encode the words “Coca-Cola”, after all.

The thing is, geeks like to pretend that the legal system is some sort of Rube Goldberg contraption, easily foiled by their unparalleled cleverness. Sadly, this isn’t the case. All the IANAL-prefixed prattling on Slashdot about quick & easy ways to make yourself legally bulletproof when the cops/MPAA/interpol come knocking are little more than wishful thinking. It’s like holding your finger an inch from your sibling’s face and yelling, “I’m not touching you!” over and over. Your parents weren’t dumb enough to fall for that, and neither is the legal system.

He’s right about the Rube Goldberg thing. As a matter of law, the fact that something is “just a number” won’t help you if you’re guilty of violating copyright law. Moreover, the position that anything that’s “just a number” should never be restricted is obviously ridiculous. I’m perfectly comfortable with restricting (say) numbers that are JPEG representations of child pornography or PDFs of sealed grand jury testimony. Clearly some “numbers” ought to be legally restricted.


But I do think that focusing on the “number-ness” of the information at issue here is a useful rhetorical tactic. Part of what gives the DMCA a patina of legitimacy is the idea that it restrict shady “hacker tools.” Of course, this is an inaccurate and unfair characterization. Many “circumvention devices” have little to do with infringing copyright, and banning them is bad for technological progress and competition. But this is very difficult to explain to ordinary voters who don’t have the slightest idea how software works.

Although this incident didn’t exactly make that point clear, it does illustrate the breadth of the DMCA’s restrictions. Sixteen bytes is long, but it’s not that long. I think most people don’t clearly understand that “hacking tools” are, in fact, information—the kind you can write on a T-shirt or sing in a song. It bears very little resemblance to a crowbar or a lock-picking kit. And expecting to control the distribution of a 16-byte number once it’s been released into the wild is rather silly.

And I think this is the point Felten was getting at when he wrote that

When I wrote the thirty-digit number that appears above, I carefully avoided writing the real 09F9 number, so as to avoid the possibility of mind-bending lawsuits over integer ownership. But there is still a nonzero probability that AACS LA thinks it owns the number I wrote.

As a practical matter, the odds that he found one of the MPAA’s secret numbers by accident is close enough to zero that we can discount it entirely. But asserting ownership of a 16-byte number is still dramatically different than asserting ownership over a thousand-byte (text), million-byte (music), or billion-byte (movies) number. Thinking about it as an integer is still vaguely plausible—you can write it on a T-shirt or poster with lots of room left over. On the other hand, while you can theoretically represent the latest Harry Potter movie as an integer, you’d need a whole library full of paper to do it. So while we shouldn’t entirely rule out laws restricting the distribution of 16-byte numbers, such restrictions deserve much greater scrutiny than restrictions on the distribution of million-byte “numbers.”

  • http://tieguy.org/blog/ Luis Villa

    Do they really, though? If that’s your criteria, they’ll just make the key longer. The issue really ought to turn on the deeper question of what the information signifies (control over the output of an entire industry, and control of the things owned and paid for by individuals) rather than the length or arbitrariness or number-ness of the number. But it seems neither side is going to get to the more important issues any time soon.

  • http://tieguy.org/blog/ Luis Villa

    Do they really, though? If that’s your criteria, they’ll just make the key longer. The issue really ought to turn on the deeper question of what the information signifies (control over the output of an entire industry, and control of the things owned and paid for by individuals) rather than the length or arbitrariness or number-ness of the number. But it seems neither side is going to get to the more important issues any time soon.

  • http://www.techliberation.com/ Tim Lee

    Well, I’m not proposing that we have a constitutional amendment protecting the right to transmit any number smaller than 32 bytes. I just meant it as a rough heuristic for thinking about situations in which the law is controlling the distribution of information. It certainly works in copyright: all else being equal, shorter excerpts are more likely to constitute fair use than longer ones.

  • http://www.copyrightings.com Kevin

    I think this civil disobedience is in the same vein as MLK Jr’s. A digital march on Washington:

    http://www.copyrightings.com/2007/05/digital-march-on-washington.html

  • http://www.techliberation.com/ Tim Lee

    Well, I’m not proposing that we have a constitutional amendment protecting the right to transmit any number smaller than 32 bytes. I just meant it as a rough heuristic for thinking about situations in which the law is controlling the distribution of information. It certainly works in copyright: all else being equal, shorter excerpts are more likely to constitute fair use than longer ones.

  • http://www.blurringborders.com kdonovan11

    I think this civil disobedience is in the same vein as MLK Jr’s. A digital march on Washington:

    http://www.copyrightings.com/2007/05/digital-ma

  • http://tieguy.org/blog/ Luis Villa

    all else being equal
    But this is exactly the exception that proves the rule. Look up Harper and Row v. The Nation :)

    I think this civil disobedience is in the same vein as MLK Jr’s.

    I think MLK is rolling in his grave at the mere thought of that.

  • http://tieguy.org/blog/ Luis Villa

    all else being equal
    But this is exactly the exception that proves the rule. Look up Harper and Row v. The Nation :)

    I think this civil disobedience is in the same vein as MLK Jr’s.

    I think MLK is rolling in his grave at the mere thought of that.

  • http://www.techliberation.com/ Tim Lee

    Luis, I was actually going to cite Harper, but decided the post was long enough already. The text at issue in that case was about 300 words long, or about 1000 bytes. I think you’d be extremely hard-pressed to find a case where someone excerpted a 16-character string and it wasn’t fair use.

  • http://www.techliberation.com/ Tim Lee

    Luis, I was actually going to cite Harper, but decided the post was long enough already. The text at issue in that case was about 300 words long, or about 1000 bytes. I think you’d be extremely hard-pressed to find a case where someone excerpted a 16-character string and it wasn’t fair use.

  • http://sethf.com/ Seth Finkelstein

    Folks, it’s not about *infringement*. It’s about *circumvention*. Your intuitions, the sense of what’s “fair use” for *infringement*, DO NOT APPLY to the all-new completely-different offense which is *circumvention*.

    I see this confusion coming up over and over in the blog reactions.

  • http://labs.echoditto.com tom

    Thanks for the link, Tim, and the thoughtful commentary. I’m not sure I agree with all of it, though.

    First, I worry about any kind of length-based criteria for granting copyright. Are visual works going to be treated as bitmaps or vector art? If the latter, some minimalist artists are going to be in trouble. Will poets start worrying about increasing the average entropy of their stanzas?

    Second, I think this quote gets at the root of the problem with Felten’s argument:

    …asserting ownership of a 16-byte number is still dramatically different than asserting ownership over a thousand-byte (text), million-byte (music), or billion-byte (movies) number.

    I think the confusion comes from the implication that the AACS LA is asserting ownership — it isn’t. You’re still perfectly welcome to use those numbers for other purposes. You just aren’t allowed to redistribute them for a specific purpose that victimizes the AACS LA. The analogy I drew in Felten’s comments is to a neighborhood association asking a store to restrict the sale of spray paint to minors. It’s not asserting ownership over the paint, just restricting its distribution for a particular use.

    With that said, I think that the idea of suppressing information because of what *could* be done with it is inherently problematic, and that the DMCA is bad policy. But I don’t think the number-ownership argument makes a lot of sense.

  • http://sethf.com/ Seth Finkelstein

    Folks, it’s not about *infringement*. It’s about *circumvention*. Your intuitions, the sense of what’s “fair use” for *infringement*, DO NOT APPLY to the all-new completely-different offense which is *circumvention*.

    I see this confusion coming up over and over in the blog reactions.

  • http://labs.echoditto.com tom

    Thanks for the link, Tim, and the thoughtful commentary. I’m not sure I agree with all of it, though.

    First, I worry about any kind of length-based criteria for granting copyright. Are visual works going to be treated as bitmaps or vector art? If the latter, some minimalist artists are going to be in trouble. Will poets start worrying about increasing the average entropy of their stanzas?

    Second, I think this quote gets at the root of the problem with Felten’s argument:

    …asserting ownership of a 16-byte number is still dramatically different than asserting ownership over a thousand-byte (text), million-byte (music), or billion-byte (movies) number.

    I think the confusion comes from the implication that the AACS LA is asserting ownership — it isn’t. You’re still perfectly welcome to use those numbers for other purposes. You just aren’t allowed to redistribute them for a specific purpose that victimizes the AACS LA. The analogy I drew in Felten’s comments is to a neighborhood association asking a store to restrict the sale of spray paint to minors. It’s not asserting ownership over the paint, just restricting its distribution for a particular use.

    With that said, I think that the idea of suppressing information because of what *could* be done with it is inherently problematic, and that the DMCA is bad policy. But I don’t think the number-ownership argument makes a lot of sense.

  • http://www.digitalproductions.co.uk Crosbie Fitch

    Seth, I’ve observed that confusion too. I think it helps demonstrate that copyright and DMCA constitute sophistry in the subconscious minds of the general public.

    You buy something, it’s yours.

    You have a device that let’s you copy it, you can.

    Someone tells you the magic number that fixes the otherwise broken copy function on your device, why not use it? Why not tell it to your friends?

    It’s your property, your device, your option to copy…

    “No, no! You can’t do that. We all agreed three centuries ago that we wouldn’t use the copy function on our devices.”

    “Eh?”

  • http://tieguy.org/ Luis Villa

    But the whole point of Harper, Tim, is that it isn’t the length that matters, it is the substance of it. Harper, in essence, says ’300 words or 30,000 words or 3 words, if it guts the commercial value of the work, it has legal significance which the court must attend to.’[1] Here we’re 10 times less raw information than Harper, but the economic and political significance of the number is vastly larger. To pretend otherwise- really, to pretend that the length has any relevance at all- is to be in denial about what we’re really talking about.

    Again, this is not to say that I approve, but again, the length and the fact that this is a number, and not ‘open sesame’ or whatever other passphrase one wants to use, is really irrelevant in anything other than an emotional/irrational sense. (Wars have been fought and millions have died over the emotional and irrational, so that is fine do discuss too — in fact I think that is mostly what Ed is talking about — but again, lets not get things confused.)

    [1] Seth, I use ‘legal significance’ and not infringement for your benefit :)

  • http://lippard.blogspot.com/ Jim Lippard

    “You’re still perfectly welcome to use those numbers for other purposes. You just aren’t allowed to redistribute them for a specific purpose that victimizes the AACS LA.”

    I would argue that most of the current distribution of the key in question is NOT for circumvention purposes, but rather as a statement of opinion about such distribution. I.e., it’s political speech.

  • http://www.digitalproductions.co.uk Crosbie Fitch

    Seth, I’ve observed that confusion too. I think it helps demonstrate that copyright and DMCA constitute sophistry in the subconscious minds of the general public.

    You buy something, it’s yours.

    You have a device that let’s you copy it, you can.

    Someone tells you the magic number that fixes the otherwise broken copy function on your device, why not use it? Why not tell it to your friends?

    It’s your property, your device, your option to copy…

    “No, no! You can’t do that. We all agreed three centuries ago that we wouldn’t use the copy function on our devices.”

    “Eh?”

  • http://tieguy.org/ Luis Villa

    But the whole point of Harper, Tim, is that it isn’t the length that matters, it is the substance of it. Harper, in essence, says ’300 words or 30,000 words or 3 words, if it guts the commercial value of the work, it has legal significance which the court must attend to.’[1] Here we’re 10 times less raw information than Harper, but the economic and political significance of the number is vastly larger. To pretend otherwise- really, to pretend that the length has any relevance at all- is to be in denial about what we’re really talking about.

    Again, this is not to say that I approve, but again, the length and the fact that this is a number, and not ‘open sesame’ or whatever other passphrase one wants to use, is really irrelevant in anything other than an emotional/irrational sense. (Wars have been fought and millions have died over the emotional and irrational, so that is fine do discuss too — in fact I think that is mostly what Ed is talking about — but again, lets not get things confused.)

    [1] Seth, I use ‘legal significance’ and not infringement for your benefit :)

  • http://lippard.blogspot.com/ Jim Lippard

    “You’re still perfectly welcome to use those numbers for other purposes. You just aren’t allowed to redistribute them for a specific purpose that victimizes the AACS LA.”

    I would argue that most of the current distribution of the key in question is NOT for circumvention purposes, but rather as a statement of opinion about such distribution. I.e., it’s political speech.

  • Doug Lay

    I wonder if the Supreme Court’s decision earlier this week in Microsoft vs AT&T may make a difference here. The court, in a 7-1 decison, ruled that a software code cannot by itself be a component of a patented invention, since the code itself is only a blueprint providing instructions for execution by a physical machine. Hmmmmmm…software is NOT a component by order of the Supreme Court. Yet the law forbids the distribution or posession of a component of a circumvention device. Hmmmmm…..

  • Doug Lay

    I wonder if the Supreme Court’s decision earlier this week in Microsoft vs AT&T; may make a difference here. The court, in a 7-1 decison, ruled that a software code cannot by itself be a component of a patented invention, since the code itself is only a blueprint providing instructions for execution by a physical machine. Hmmmmmm…software is NOT a component by order of the Supreme Court. Yet the law forbids the distribution or posession of a component of a circumvention device. Hmmmmm…..

  • http://enigmafoundry.wordpress.com/ enigma_foundry
  • http://enigmafoundry.wordpress.com eee_eff
  • http://www.manifestdensity.net tom

    I would argue that most of the current distribution of the key in question is NOT for circumvention purposes, but rather as a statement of opinion about such distribution. I.e., it’s political speech.

    That’s an interesting point, Jim. I’m inclined to agree, and am very curious what a court would say about it.

  • http://www.manifestdensity.net tom

    I would argue that most of the current distribution of the key in question is NOT for circumvention purposes, but rather as a statement of opinion about such distribution. I.e., it’s political speech.

    That’s an interesting point, Jim. I’m inclined to agree, and am very curious what a court would say about it.

Previous post:

Next post: