X-Message-Number: 1770 Date: Fri, 19 Feb 93 11:47:56 CST From: Brian Wowk <> Subject: CRYONICS Uploading It's interesting that Raymond Kurzweil has discovered uploading, but like Michael O'Neil I think his timescales are way off. I'm a great fan of Moore's Law (exponential increases in computing power with time), and I think that optimism in this area is well justified since we are still far from the physical limits of computer miniaturization. Increasing parallelism alone will be sufficient to support Moore's Law indefinitely. However, Kurzweil, like most "uploaders" knows more about computers than about medical imaging or neurobiology. There are fundamental physical obstacles of dose and scanning time that prohibit x-ray CT from ever achieving micron resolution of an intact human brain. (This question was beaten to death on the net last year.) Things are not quite so bad for MRI, but they are not good either. Over the last decade MRI engineers have lowered the noise in the electronics to the point where the limiting noise source in MRI is now Johnson noise (or thermal noise) in the patient. Even if a surface coil is used to limit reception to the brain, the signal from a single cell deep inside the brain will be swamped by noise from the rest of the brain. (Yes, MR images of single cells have been obtained, but that is because the cell- sized coil is placed right next to the cell being imaged.) In short, there is no foreseeable means to read out memories from a human brain non-invasively. This job will require sophisticated nanomachines ("mature nanotechnology"), and is therefore at least 50 years away from being doable. It's nice to see such vision and optimism in the computer science community, but we have to figure out how to redirect this enthusiasm back to the only real-world prospect of high-tech longevity: cryonics. This might not be as easy as one might think. Several years ago I was floored when I read what Hans Moravec thought about cryonics when asked in Omni magazine. He essentially said that cryonics was "too crude" to interest him, and that he was much more fascinated by the prospects of hyper-advanced technology reconstructing a deceased person's identity from clues left in the world around them. (Of course, he is contradicting himself: What better clue could a "deceased" person leave as to their identity than their brain!) After reading Hans Moravec, the quintessential uploader, dismiss cryonics as too crude, I really don't know what to think about uploaders anymore. Most of them will probably die along with the rest of the mundane crowd, with great irony, from lack of vision. --- Brian Wowk Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=1770