X-Message-Number: 7856 From: (Thomas Donaldson) Subject: Re: CryoNet #7848 - #7853 Date: Sat, 15 Mar 1997 01:01:49 -0800 (PST) Hi guys! Since my connection broke and its now past my bedtime, I shall try to be brief. 1. We must first understand how our brain works much better than we do now, and without that understanding the information we might get from any kind of readout will mean nothing. Not only that, but until we understand we won't even have a clear enough idea of what information is important and what is not. But then: that understanding of how our brains work will bring with it also a much better understanding of how to fix them. The two go together inevitably. 2. Any serious proposal to read out brains will also have to be automated: no one can be expected to do all the cutting and slicing by hand, and keep track of it by hand, and so on and on. Not only that, it will have to get out a VERY ACCURATE and COMPLETE version of our brain. We may not even know that it is inaccurate and incomplete until we have the "person" read out --- and find that he or she is implausibly faulty in many obvious ways. It must be especially accurate and complete because there will be no second chance: the readout, as presented, is destructive, and we can't go back for a second look. This requires a very high standard of operation if we claim to bring someone back by using it. I can think of ways to readout brains which are nondestructive, too. But along with them comes a much increased ability to repair the brains in place. 3. Some people seem to think that the clunky machines we have now and call "computers" can do all kinds of wonderful things. We do not know now even what the people of 50 years in the future will call "computers", if they even use that word, or what their principles of operation may be. I would not want to become a program in any present "computer". In a sense I am already a "program" in another kind of computer; take care, then, when you blithely talk of reading in and reading out. Perhaps the computer you will find yourself in will look like a brain. As for whether our "computers" can do wonderful things, that depends on where you stand. Even giving one lots of memory and lots of processors you won't have something which can really imitate a human brain. In some ways, sure, it will be better, and in others worse. But I doubt it can be done with anything like our present "computers". (Yes, I know about computers, particularly parallel ones --- it was my specialty. And I've done a lot of study of brains, too, because they are important for cryonics). Could we emulate a flatworm's nervous system. Yes. Probably even that of a snail. But a human being? Not at all --- I doubt that we could even simulate the brain of a mammal, bird, or reptile. Maybe a few very stupid fish. It's not just a matter of number of processors and memory, it's a matter of the kind of processing which needs to be done. Do I really claim this is a sufficient answer and you guys will be silenced forever more? No. But it's all I'm saying tonight. Goodnight, and long long life, Thomas Donaldson Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=7856