Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!rutgers!aramis.rutgers.edu!athos.rutgers.edu!nanotech From: mike@everexn.com (Mike Higgins) Newsgroups: sci.nanotech Subject: Re: Nanotech Economy Message-ID: Date: 13 Dec 90 21:58:36 GMT Sender: nanotech@athos.rutgers.edu Organization: Everex Systems, Inc. Lines: 56 Approved: nanotech@aramis.rutgers.edu In erich@eecs.cs.pdx.edu (Erich Stefan Boleyn) writes: > Well, I think a better question to ask would be how would you "expand" your >brain in size? And *why* would you do it? Before you answer me with flames >(;-) I'd like to say that as a student of neuroscience and genetics, I've been >thinking about this for a while, and the brain is so specifically wired (so I too have been thinking about how to increase intelligence, but I look at individual systems, and assume a gradual improvement. Let's find some engineering tasks we can tackle first, and worry about what 'intelligence' means later. Doing the engineering will force us to learn all the small details that will make the whole picture obvious one day. For example: How about an image processor on the optical nerve? We could design a machine to 'infect' the optic nerve and replace a piece of it with a processor. Normally, the signals would be passed through unchanged, but pictures could be constructed, enhanced, and sent down the nerve like instant replays. You could see in lower light, find edges, detect motion, do Fourier analysis, cubic interpolation zooms, and much more. By growing a few gigabytes of storage you could have a photographic memory. Where do you find space for this processor and memory? Biological systems are often very inefficient: the optic nerve sends constant signals when exposed to a black image! The brighter light you see, the fewer ions the nerves send. You could put an inverter at both ends and run electrons across the middle. This would remove the necessity for all the organelles in the middle that used to support all that black signal current! As time went by, I would want more from my optic nerve processor: Why waste processing power doing cubic interpolation, how about growing up the optic nerve and replacing all the rods with 6 solid-state detectors each! Now I can do a 6 times zoom without degrading the image. The processor on the nerve constantly down-samples the image to feed it to the old visual cortex in the format it is used to. What a waste! So lets grow down the other end of the optic nerve and 'watch' the visual cortex in operation. We won't know how it works when we start, but we sure will by the time we find a way to replace it! How about growing a combination speech_recognition/speech_synthesis/ high_speed_modem/celular_phone in your head? We could almost build that now without nanotechnology! The speech recognition system listens to you sub-vocalizing commands, while the speech synthesis system whispers in your ear. You could call up medical databases, run expert programs on other systems, feed symptoms to it and listen to the diagnosis. (Remember, he's not a REAL doctor ;-). How 'intelligent' is a human that has a photographic memory and can 'recall' any fact stored in any publicly available database? Not to mention doing all the boring calculations localy. For the symbolic integration, you would call up your PC at home and run Derive. (Well, that's how I do it now, that sort of processing will be smaller soon). Of cource some 'native intelligence' (whatever that means) is still necessary to set up the problems. But I think a person equiped with these devices in their head would break the scale on any IQ test currently known to man. These two additions are simple things, and parts of them already fit in your hand or in your lap. I imagine that after a few years of tinkering with systems like this in your head, you would know how the brain used to work. Used to, because it no longer works that way: it's all been replaced with silicon, nanocomputers, quantum computers or whatever. mike@everexn.com