From: David Picon Alvarez (eleuteri@myrealbox.com)
Date: Sun Oct 30 2005 - 05:54:29 MST
> 'Flash' uploading, involving scanning the brain and copying the data to
> a computer, and Moravec transfer, involving replacing individual neurons
> with functional equivalents, are not the only possibilities.
There is another possibility, which I first read about in a Greg Egan sci-fi
story, _Closer_, and which I think would deal with all the issues brought
up.
The idea is that, as it has been claimed, the brain is a dynamic machine,
and recording static information might not suffice. So instead of
deep-scanning a neuron and replacing it with a mechanical component, a
neuron is "isolated" and studied by a system that transparently keeps it
connected to the rest of the brain while it studies the way it reacts to
inputs. After a while, considering the knowledge we would have at the time
of the internal structure of the neuron, we could build a dynamic model of
the neuron, and replace it with a mechanical sync'ed component that keeps
dynamic state. The way to do this proposed in the book is by non-invasively
carrying forward this process while writing the dynamic model on a computer
substrate, and after the model is complete and has been compared to the
biological original for a while in order to ensure that there are no errors,
the biological components are discarded altogether.
This would deal with the dynamicity issues mentioned above, as well as the
matter of so-called pressence, since the dynamic state of the mechanical
substrate would share all the structure and causal links with the biological
substrate.
--David.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT