X-Message-Number: 15008 Date: Fri, 24 Nov 2000 06:48:13 -0500 From: Thomas Donaldson <> Subject: yet more: computers. brains. etc This is for Mike Perry, who remains the only person so far on Cryonet who has tried to seriously answer my question. 1. Yes, I consider timing very important. After all, it IS very important in our own lives. Any version of a "human being" that could not work in real time is basically going to be an imaginary creature. 2. I also came up with several other features not presently shared by any existing computer neural net. Since you read PERIASTRON, you'll know of work which shows that the phenomena I described (swift changes in connectivity and production of new neurons) really do happen. Given the timing problem, it's still far from obvious to me that such a neural net can be easily imitated by a computer. I hope this at least clarifies the problem in terms of what it might mean to produce a "human" device. I am not insisting that anyone make real hardware; I am asking instead for information about how they propose to make such hardware... in other words, for a theoretical solution (using real things, not totally imaginary devices). There seem to be various ideas about brains going about on Cryonet. For what it's worth, my questions here have nothing to do with quantum mechanics. I am proposing actual features of brains which may make it impossible to consider them as Turing machines, rather than ABSTRACT possibilities which might have that effect. They can, I believe, be explained very simply: yes, in one sense of the word, computers are intelligent (not self-aware, but intelligent). If we knew that any form of intelligence must necessarily be imitatable by a Turing machine, then the conclusion that brains are Turing machines would come out automatically. But is that really true? Especially since time plays a large role in our thinking and behavior, and we do NOT simply sit and think (even if we're smart theorists we still eat, sleep, and deal with other events around us), must we still be Turing machines (at least in the sense that we can be mapped into them?). And if timing alone is sufficient to make that impossible, we'd have an interesting result, too. Best wishes and long long life, Thomas Donaldson Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=15008