Ivan Seeking said:
we can't yet quantify consciousness or self-identity.
No more than we can quantify the operations of a complex artificial neural network, but we can sensibly conclude that this is where the operations reside.
Ivan Seeking said:
there is no way to say if a person's consciousnes could ever be transferred in such a manner
If my above reasoning is true, then it necessarily goes to show that this is, at minimum, theoretically possible.
Ivan Seeking said:
We could never know if it really is good ole Joe, or just a fascimile.
If Joe and Mr.fascimile are equivalent, then it is also necessarily true that Joe remains Joe.
Ivan Seeking said:
We have the same problem with the notion of Star-Trek-like transporters.
It's a misconception not an unresolved problem. Although I suppose the fact that it's a misconception is a problem in itself. If the function of the mind has been transferred, the mind, by implication, has been transferred.
Ivan Seeking said:
How could any process of uploading produce the mind of the original unless the original is deleted in the process?
The upload could occur slowly, shifting data to new neurons at more or less the rate at which a biological brain does it. It's deletion just the same, as is the natural working of the brain, but once again, if the reproduction is equivalent this is inconsequential. The rate of transfer is also inconsequential, even if practically instantaneous.
Ivan Seeking said:
so you really couldn't be sure if you are you or just a copy, except through the circumstances.
In any logically consistent way you could possibly look at the problem, if the copy is a copy of you, then you are you.
Evo said:
What if you live beyond death but have so senses, no sight, sound, touch, etc, you have no body, no one can see you or sense you in any way and you can't communicate. But you are conscious. Not too practical, eh?
Not a particularly pleasant state of existence, but I see no reason for the requirement that an uploaded individual will be stuffed into a dark black tactile-less box. The transfer is the hard part, once that is over with the person could be provided with an existence which makes the former reality appear pitiful and depressing.
There is no apparent restriction for making such a transfer. Molecular resolution scans of biological structures are required, which are possible today but extremely impractical for such a high volume of matter. Voxel data interpreters for neural network reconstruction also exist. This is quite recent. I recall 1 cubic mm teravoxel data of biological scans being used for research at the moment with function being largely limited by processing power.
And that is the second problem. Processing power. There are estimates for functional recreation around. IBM if I recall correctly expects a 1 Exaflop requirement for tangible electromechanical equivalency and something below this for reduced-model neurological equivalency. Markram's chronological projection is 2018.
http://www.philosophy.ox.ac.uk/__data/assets/pdf_file/0019/3853/brain-emulation-roadmap-report.pdf"
And there are estimates for molecular-precision models and even far fetched quantum models. These are very improbably a requirement for functional recreation, however nevertheless a potentially useful physiological reconstruction.