Tom, As I make clear in my paper, having a high algorithmic complexity does NOT mean that an algorithm is intelligent. As you point out, a large algorithm full of random junk that does nothing has high algorithmic complexity and clearly is not intelligent.
What I say is the reverse, that if an algorithm is very "intelligent" (actually what I say is that it is a very powerful predictor in a sense that I precisely define in my paper, read it there if you want the details) then the algorithm must have a high algorithmic complexity. When Matt says that I use algorithmic complexity to define intelligence, he means that I have used algorithmic complexity in my definition, not that I have simply equated the two... which clearly is not enough as your example illustrates. Does this help clear things up? Cheers Shane ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=4007604&user_secret=8eb45b07
