"It's not even correct to say "the probability of this being from a random generator is very low" because that's not true - it either was or was not randomly generated, that's not open to probability."
This isn't true when you take a Bayesian approach, where probability is a statement about your knowledge rather than reality. In fact, it seems like the whole issue could be greatly simplified/rationalized through application of Bayesian ideas - "Color" then being determined probabilistically (in the Bayesian knowledge sense) rather than absolutely.
Perphaps I'm misreading you, but isn't that directly against the main gist of the article. Historical events happened in a certain manner, there is no probability involved. It's all 1.
Presented with partially unknowns about historical events one can use different approaches to decide how to proceed, but when talking about law, most things revolve specifically not about what's probable, but what actually happened. There is no "probable". Isn't that the main point of the article?
It's quite easy, from a law perspective, to decide what to do when no sure about what actually happened: let the accused go.
For most of us, though, this is not the type of answer we're after, and that's one of the things the article is trying to make us understand: our perspectives are different.
This article is an oldy but goody about intellectual property law and how programmers, et al, routinely fail to understand it because of a paradigmatic difference in how they view data versus how other people (and the law) views information.
Copyright makes the act of copying something illegal. The only thing the bits might give you is evidence of copying. Trying to think about the law governing human actions only in terms of bits unsurprisingly fails.
No, the copying is illegal as well. It's just that copying without redistribution is typically both hard to catch and not worth the copyright holder's attention.
This isn't true when you take a Bayesian approach, where probability is a statement about your knowledge rather than reality. In fact, it seems like the whole issue could be greatly simplified/rationalized through application of Bayesian ideas - "Color" then being determined probabilistically (in the Bayesian knowledge sense) rather than absolutely.
Interesting article though.