"Even assuming it were just a set of energy states in the brain, and it isn't, brains aren't digital. Think of digital audio with the ten layers of white noise..."Actually you just make the noise more subtle. It is—in principle—impossible to eliminate noise when converting between analog and digital. In practice, you can make the noise be composed of so many pixels ("high resolution") that people don't notice it, but, well, "little inaccuracies nobody notices" may be just fine for music or photos, not so much when it comes to people's minds.
You can always, in principle, get rid of the noise by digitising/simulating with sufficient resolution, at increasing cost in memory and computer resources. It's possible that a decent brain upload and simulation would be possible but restricted only to the rich, and everyone else would have to make do with something low-res and glitchier.
Leaving to one side that according to the Lucas-Penrose argument in cybernetics, the mind is unrepresentable in machine logic—that's what I meant by "it's not just a set of energy-states in the brain". Human beings can construct Gödel propositions, machines are—by definition—incapable of doing so. And all the counterarguments to Lucas-Penrose are merely exercises in utterly missing the point.
Sorry, mind-uploading makes FTL look easily feasible. FTL only violates our current conception of the laws of physics; mind-uploading is a logical impossibility.
1 comment:
D'oh, my apologies.
I do not (in the absence of strong arguments) buy the idea that small errors are unacceptable when it comes to minds - it is not obviously the case, and the disruption of the brain represented by digitisation errors could conceivably be less than physical disruptions that the form of the mind is known to survive, like drink, drugs and sharp blows to the head (....well, it doesn't always survive intact, but there's usually some continuity of identity at least.) Unless you're thinking of error in computer simulation of the individual parts of the brain (neurons, synapses, etc.), which would be a more serious matter.
(Which as you note is all irrelevant if the Lucas/Penrose argument holds, but I haven't thought about that sufficiently to comment.)
(Also, he rather than she.)
Post a Comment