X-Message-Number: 33038 References: <> From: Gerald Monroe <> Date: Sun, 7 Nov 2010 06:30:14 -0600 Subject: Re: CryoNet #33036 - #33037 --00163630f0e993bb92049475ac86 Perry : for the repair task to ever complete it has to happen faster than cosmic rays damage what you are working on. There is such a thing as "too slow". Also, how could you test or validate a process that takes a century? Even if you could construct "free floating" nanomachines capable of operating while immersed in liquid nitrogen, how would you get the data out of them? You'll need a gigantic computer with enough memory to store a compressed form of an atomic map of every single atom. (well, you probably would run compression filters at the very start of the process on the incoming data, but it would still need ridiculous amounts of memory) Nanobots working alone probably cannot have enough intelligence to make meaningful repairs within the laws of physics. And THIS law is a basic assumption : that you cannot make a computing element out of anything smaller than clusters of a few atoms each) When I say "gigantic", I mean in capabilities compared to the current state of the art. With molecular circuitry, which you would have the ability to create by definition if you had molecular manufacturing, the computer might fit in a closet - but it would need to be supplied with many watts of cooling and power. (unless you could somehow make "reversible computing" work but that only works with limited types of algorithms) Anyways, here's the "base assumption" that a working device might resemble. It's how I visualize it, I'm aware there's probably many other ways that might work better. The machine's a flat plate, about the size of two human heads and a few centimeters thick. It's probably all the same color on the outside because it's entirely composed of identical repeating blocks of nanomachinery. On the edges there are all sorts of data and power and coolant connections. At the molecular level the device is composed of many many trillions of nanoscale parts that can remove chunks of tissue a few atoms at a time. The patient is still at the same temperature they were stored at, but the machine might operate at higher temperatures internally. Internally the machine can analyze these chunks to determine exactly which atoms are in each and then the machine then pipes the along along a highway of molecular conveyer belts to the other side where it reassembles the brain a chunk at a time. It also reports to the control computer a compressed version of the atomic structures it discovers. Software determines from analyzing the orientations and densities of the proteins and signalling molecules the 'state' of each synapse in the brain. It then corrects the damage by substituting in to the molecular 'printing pattern' the templates for cells that we know from experimentation are "known good", except that these cells will have the consensus DNA sequence of the original patient and the synaptic states of the original patient. These cells will have to survive thawing, so they would probably be reinforced or have temporary support nanomachinery that would mostly remove itself after you warm up the brain and have it running again. You could also insert "boot up code" into the genomes of each neuron that would cause certain changes to be made after warming, to further put the neurons closer to the original condition they were in when the person was alive. The computational algorithm you would need to do this process would probably perform many complex decisions that require knowledge of large sections of the brain's structure. Repairing a crack, for instance - said crack would cleave many axons and require some kind of simulation involving billions of atoms to calculate what the original structure prior to freezing was. No way a bot composed of a few hundred thousand atoms and stiff from low temperature could figure out how to correctly fix a crack, unless there are properties of matter we are unaware of. Not to mention - how would the bot eliminate the waste products it would produce? Any energy-using device must emit waste of higher entropy than the starting materials. Not a good idea to poop out waste while floating in the middle of a brain you are trying to fix, whether that waste is heat or chemicals. This process, in order to create a working brain that is not afflicted by all the biochemical changes caused by death, will have to make many many changes. Some information will be lost. For ethical reasons, we could do the process in 2 passes through this machine (the first pass gets the atomic map but just creates the brain again like it was and the second rebuilds everything with healthy neurons). We could also store a copy of how the brain was originally so that as technology advances more information could be extracted from it. I think the most accurate revivals could be done by using the synaptic mappings to create emulated minds that run on molecular computing hardware at millions of times faster than current human thought. I would guess that if you could even poorly copy a few scientists this way, they could solve the more complex problems very well with a million years of time to consider the problem for every year on earth. These super-intelligent beings would probably be able to fully extract nearly all of the information contained in an atomic map of a person's brain, and make a version of the person indistinguishable from the original. Regarding cost - building a single instances of one of these machines would be cheap. While an end use device like this might be too specialized to self replicate, molecular manufacturing equipment will be able to copy itself, making additional copies cost very little. All the cost is in building copy number 1. This is also why it probably won't matter much for those rich folks considering cryonics whether their wealth is kept in trust to be 'used only for them' or 'diluted' among all of the patients. It's also hard to imagine how standard investments would hold their value in a world that has molecular manufacturing. This is also why putting the majority of the funds into the Patient Care Trust for keeping the patients frozen makes the most sense. In a century from now, when molecular manufacturing is available and it is obvious that repair of cryonically frozen humans is practical, there would be enormous pressure to develop the technology. All those billions of people living then would want medical care that could bring them back no matter what. With any luck, after developing the tech for common use they would use it on those dusty old cryostats containing their sloppily frozen ancestors. (by the standards of a century from now) --00163630f0e993bb92049475ac86 Content-Type: text/html; charset=ISO-8859-1 [ AUTOMATICALLY SKIPPING HTML ENCODING! ] Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=33038