X-Message-Number: 8586 From: Thomas Donaldson <> Subject: Re: CryoNet #8541 - #8545 Date: Thu, 11 Sep 1997 23:02:38 -0700 (PDT) Hi again! Yes, these comments are somewhat late. As for the notion of equivalence, I do hope that John Pietrzak read and listened to what I said. There is no single definition of equivalence, there are many depending on what you want to do. The ideas of Turing about computers use a broad definition of equivalence. Practical issues raise the validity of much narrower definitions. I come from a math background, though now I've worked with computers and computer scientists for some time. I've also had occasion (I've been seriously involved with cryonics for a LOOONG time) to read and think a good deal about biology. And thus: 1. No, our brain is very unlikely to work by quantum logic or other means suggested by various physicists. As I said, we evolved to an optimum, and are still evolving (yes!). Immortality may change the situation, though --- not by doing away with evolution, but by making human beings no longer the units of evolution. Competition would be not between human beings but between specifications for human beings (right now, genes). 2. This issue of optimality continually seizes those who haven't thought much about biology. For instance, so far the devices (such as they are) implementing quantum logic will only work at very low temperatures. Keeping temperatures that low on the Earth requires lots of energy for refrigeration. Life forms will of course eventually occur in interstellar space (probably evolved from human beings) and the setting will be different. If quantum logic brains become possible, then that would be their most likely location. To keep one working here would require too much cost for the output. We became highly parallel instead. (I do not mean here that human beings might not make and use them. Just that they would not have EVOLVED HERE because they are in no way optimal for the processing an animal needs to do). The same may be said of semiconductors, superconductors, etc etc. It is not enough to simply show that on one parameter a computer system (or any device) is superior to anything biological. You have to consider the setting too. The materials and energy which make our brains are far less expensive than those which went into making the specialized chess playing computer which beat Kasparov. Carbon, hydrogen, and oxygen are all over the place. Iron is the most common metal. The other constituents occur only as traces: otherwise, they would take far too much effort for the return. To say that a biological system might not reach a "true" optimum by evolution is not a claim about biology at all. It is a claim about genetic algorithms. If we are trying to find a minimum or maximum in a field where we know very little about the function we are trying to minimize or maximize, a kind of genetic algorithm has lots of merit. We pick random locations and try to increase from there, using the largest ones. Given that we know very little about the function, a genetic algorithm has lots of merit. The function may be continuous but not even differentiable, for instance --- many things in the real world aren't differentiable, after all. And the totality of life on Earth provides a lot of parallelism, too. I have got a few books on how to optimize: not just linear programming but various other such issues, and they are quite respectful of this method. Naturally, if we DO know about the function then there are much better methods. Personally, I think we have a lot to learn from biology. STILL. It's the only kind of nanotechnology presently operating, among other things. 3. The original Turing test, with limited means by which the interrogator could even talk with the computer/person on the other side, fails because it does not take in the full range of behavior a human being can show. Issues of intelligence and its meaning, while I certainly agree that they are far more vague than most people think, aren't the central problem with it. The central problem is that it operates only with symbols. Yes, folks, deep down our brains do not work with symbols. We operate with our perceptions, and possibly because we are human beings we've evolved brain areas specialized to deal with language. But those areas are useless without the rest of our brain. Nor can they be identified with thinking: when we use symbols, we know their MEANING, which ultimately cannot be given with other symbols. It's one thing to be able to natter on, say, about plants, and quite another to recognize them and their workings on a microscopic slide or walking in the hills. Neither of these abilities is tested by the Turing test; yet if you lack them you really know nothing at all. I've said before in this forum that I do not doubt that we can build a device capable of whatever a human being is capable of... and not by the ordinary means. Whether you will consider such a device a computer or a computer with peripherals depends a lot on just what you think is a computer. For that matter, human beings (and other devices like them) will be subject to the Turing limits. But then, as I said, whether that means we are computers depends a lot of just which test of equivalence you choose to use. Best and long long life, Thomas Donaldson Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8586