The author points out that bits do not actually have colour, they can only represent colour. This is the main concept that most humans struggle with because most human beings are not colour-blind like the computer science people are trained to be. This is why their is some hostility between computer science people and lawyers. Lawyers cannot comprehend that bits do not have colour at the fundamental level, and the computer people can. The author gave an example about a websight that shows when the sun rises and sets. The catch is that the websight has a disclaimer that says that the information is not suitable for court. I was completely baffled when I read this, why would that be? The author answers this by saying that in order to actually know when the sun rose or set, one would need an expert witness to prove it, because the bits mean nothing and can be infected to give faulty information. What you need are those bits, coloured with the colour that allows them to be usable in court. This is what the witness does - he provides the reliable source to those bits. This makes whole concept of where the bits came from is just as important as the bits themselves, if not more.
Also, this article made me question what my definition of random is as well. I previously believed that random meant anything that didn't have an expected result. However, since it is important to know where the bits came from (and in this example, the numbers), then when I say a random number, what makes it truly random? I now understand that what makes that number random is not the number itself, but the way the number was generated - the colour of the number.
I disagree with the author when he says that the distinction of colour should be considered when child pornography is the topic. I think that it is dangerous by its very existence - the danger is in the bits, not the colour. My reasons for thinking this have to do with my moral values, which might be a missing key element when most people consider a topic such as this.
I also liked his sunrise/set example. He did a really good job of showing the dichotomy between laws and computer science, but I wish he had shown more in the way of reconciliation. I'm like that with everything I read; sure, the author can point out flaws, but I also wish he would point out solutions.
ReplyDeleteGood comment on randomness, it is an odd way to look at it, when considering it's contemporary conventional usage.
I agree, he clearly listed all the problems and issues, which only takes a mediocre writer. What would have made his article better is if he had presented some sort of way to resolve the issue between lawyers and computer scientists. He said that lawyers need to understand that bits don't have colour, and that computer scientists need to learn to view bits with colour for the purpose of the law, but I don't think that's enough.
ReplyDeleteYou guys realize you;re asking this guy, who has whatever his day job is, to go from a fairly insightful piece exploring a difference in point of view, to solving a problem that is profoundly hard. If he did not write until he had a solution he would probably never write. There is value in recognizing and explaining the source of a problem. It's probably fair to say comp sci types understand color as well as anyone else does -- we grow up with it in the analog world and learn a distinction. So to an extent he does suggest a solution -- have policy types learn about the world they want to regulate so they can quit making up "solutions" that make more problems than they solve.
ReplyDeleteThe contemporary usage of "random" is ... grating!