Hacker News Reader

Comment by fserb | Hacker News Reader

I think the argument that "how can a number be owned or its posession be illegal" is mostly bogus. One could translate any string of text into a series of bits and back to a single number and say that a book or a picture is just a number. Then one could say that noone invented that number, and that it's just a number, so you couldn't have copyright nor protection because everything could be randomly generated, etc, etc...

I think the real issue is that in the case of prime numbers made "illegal/protected", the absurd is not that they are protected, but that they are not encoding any meaningful information that a human being produced. They are a particular number (probably randomly generated) and the fact that they are used in a particular context (to encrypt data) makes them protected.

View on HN

The argument is bogus, because it's not the number itself that is illegal: it is the number plus the necessary meta data required to turn the number into something that has been outlawed.

It is like saying that it is silly that we prosecute people for possessing television sets, when the case under consideration is someone possessing a television set previously in the possession of someone else. If you do that, you are hiding contextual information that changes the meaning of what is being said.

As in the case of the television set, someone would only be convicted of 'possessing' or 'distributing' the number, if it was beyond reasonable doubt that he intended to break the law using the number. The number itself doesn't break the law: the actions/intent of the user is what breaks the law.

it is the number plus the necessary meta data required to turn the number into something that has been outlawed.

Where in the DMCA does it say that? And even positing that it does (though I'm pretty sure it doesn't), would that mean my possession of this prime, a linux distro, and this article in my browser history constitutes a crime?

The number itself doesn't break the law: the actions/intent of the user is what breaks the law.

Again, you're not talking about the actual DMCA. The actual DMCA implies that the intent of the user may be irrelevant if the primary purpose of the thing itself is to circumvent DRM tech. That's the whole point: the law is so poorly written and broadly applicable that nobody really knows what you can and can't do under it.

Where in the DMCA does it say that

Where in the law does it say that you aren't allowed to stab someone with a screw driver? Nowhere. That doesn't mean it isn't strictly implied.

Secondly, the DMCA isn't of particular interest here: the number might as well constitute a secret pertaining to national security, distribution of which may get you the death penalty for espionage.

The DMCA is of particular interest because it's a potential counterexample of your assertion that other tools or intent are necessary to make distribution of a number illegal. The whole point is that it's a bad law, primarily because it's unclear what it prevents you from doing.

The law is pretty clear, though, on stabbing people and committing espionage.

"but that they are not encoding any meaningful information that a human being produced"

I was sort of with you up to that point; I don't agree, but I was with you. However, now I would say you need to read the article more carefully, as the entire point is that they do indeed encode something human-produced and usable. If you use Unix, you've almost certainly got the decoder for that number sitting right on your disk right now, it doesn't get much more concrete than that. Without that, you're missing the whole point. They aren't randomly generated at all.

Guess I might as well spell out the other objection I have, which is that you're subtly begging the question. You prove that the idea of owning a number is silly because the idea of owning a number is silly in your last sentence of your first paragraph. Let me give you a much better argument: flip your argument on its head, and say that of course it's perfectly sensible to own numbers.

All but the smallest works translate in any reasonable encoding scheme into enormous numbers, numbers that would simply never be chosen randomly. There 0 chance that sitting here and stringing together "random numbers" will ever produce anything like a movie or picture or even coherent sentence, that doesn't already exist in the encoding scheme. (Not literally 0, depending on the circumstances, but close enough.) Therefore, the actual problem is that it does indeed make sense to own numbers; in an infinite space when we are all choosing umpteen millions or billions or trillions or well beyond that of digits, one can very plausibly stake an ownership claim on the grounds that it is clear that if somebody else ends up with the exact same number, they got it from you.

Therefore, since it is sensible to own numbers, the objection that it is not collapses and then the rest of the argument against ownership does too.

As I said, I don't actually agree with that line of thought. There are a lot of intricacies in the issues of encoding, and some other objections I'm leaving out, but it's still a stronger line of argument. My feeling are more related to the classic bit color essay: http://ansuz.sooke.bc.ca/lawpoli/colour/2004061001.php If that doesn't seem related, think more; it is, it very much is.

yes it does represent something human-produced but the number is not said to be illegal because the human-produced thing is protected. The number is deemed illegal because there is a convoluted way in which it could be used illegally. My point is that the protected entity is a very abstract thing much disconnected from the number itself. And therefore it does not make sense to protect it.

Regarding your second point, it is obvious to me that it is related to the bit color issue. I was trying to make a similar point to yours, but maybe I failed. Let me try again: I don't think that making this rationale of "oh, it becomes a number therefore it can't be owned" is a proper argument. Exactly because anything can become a number and we do acknowledge properties to things (like a text of a book) that would be ridiculous for a pure number to have.

My point was that even though the "just a number" argument is bad, there are other points to be made on the stupidity of this particular protection.

> you're subtly begging the question.

You are the first person to correctly use the phrase "beg the question" in something I've read in the last three weeks.

Congratulations!

The provision in the DMCA that this addresses says you can't make or distribute anything that is designed or primarily intended for breaking DRM tech.

A DVD decryption algorithm obviously is designed and primarily intended to circumvent DRM, and a prime number obviously is not. The fact that the former can be expressed as the latter effectively means the provision is either absurdly lenient or absurdly draconian. It's a law that leaves you in the dark as to what you're actually allowed to do without the threat of prosecution or legal harassment.

It is illegal to distribute a "device"(very broad definition) that can circumvent DRM.

The protest is against how broad and vague that prohibition is and that it prohibits items even if their normal/much more useful function has nothing to do with DRM. Such as prime numbers or Sharpie markers. A "Hey legislators, in your scramble to sell out to content industry you've just made these numbers illegal." kind of thing.

"One could translate any string of text into a series of bits and back to a single number and say that a book or a picture is just a number. Then one could say that noone invented that number, and that it's just a number, so you couldn't have copyright nor protection because everything could be randomly generated"

You could be on to something big there, maybe a new kind of science, even.

On behalf of Stephen Wolfram, "Ouch".

Except in NKOS-land, Wolfram has copyright on all the numbers that you might possibly ever generate...

And of course any numbers that can be derived from numbers that you might generate.

I think Wolfram should get it over with and simply copyright { Z }, that would at least make it clear where we stand on this.

Reply on news.ycombinator.com