X-Message-Number: 3871
From:  (Thomas Donaldson)
Subject: Re: CryoNet #3698 - #3700
Date: Fri, 17 Feb 1995 18:09:33 -0800 (PST)



Hi again!

To Mr. Clark:

Sorry, I must apologize. I did not make my statement that neurons were not
black boxes to one another sufficiently clear. I do agree that for simulation
purposes we can best proceed by finding SOME units which are black boxes; 
I was questioning whether or not neurons really serve well as such black
boxes. One feature of a chemical is that it diffuses over a particular area;
this means that neurons close to one another will all be affected, and in
that sense will "know" something about the internal traits of the other. There
are also pores in cell walls which again can pass chemicals between them. You
are not a black box to me because we are both human beings and I am justified
in making several claims about your internal states, EVEN WITHOUT ANY SPECIFIC
COMMUNICATION FROM YOU. It seems to me that that the reverse of that feature
is required if we define "black box".

One further comment about your postings, which I hope you will amend: our 
brains are not "goo". If they were, then neither of us could exist. They are
highly structured bodies of matter which act, in at least some respects,
as a variety of computer. Just because you don't (neither do I!) understand
fully how this particular variety of computer works does not mean that it
has no structure ie is "goo". True, the materials of which it is made,
and many of the principles by which it acts, are not those most computer
scientists are familiar with. To call them "goo" as part of a discussion of
whether or not a computer (of what kind, materials, and design?) could
emulate them is a sign of the worst kind of ignorance, that which refuses
to learn more.

When I argue that uploading MAY not work (except into computers very similar,
and possibly into ones made from material which follows the behavior of 
the material in brains very closely) I am arguing about whether or not a 
program working in one computer can be easily, obviously, or even ever 
transferred to another, different computer. In my own experience with 
parallel computing, that simply isn't true, but then the obvious rejoinder
is that the programs and the computers are different, which they would be.
I too do not believe in a "self" independent of the structure of the 
matter in which it is "contained"; I too would like to improve myself,
though I suspect that we disagree in finding the question of what is an
"improvement" more complex and difficult (for me) than it seems to be for
you.

In any case, I apologize that my last posting may have not explained my
point very well.

			Long long life,

				Thomas Donaldson


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=3871