X-Message-Number: 15030
Date: Sun, 26 Nov 2000 11:11:11 -0700
From: Mike Perry <>
Subject: Reply to Donaldson

Thomas Donaldson, #15026, says:

>A short reply to Mike Perry's answer to me on msg #15017:
>
>If your proposal to make a machine which imitates (say) my brain on a
>quantum level ie. the behavior of every atom which makes it up, then
>it's far from obvious that anyone could make such a machine.

This machine I propose would not be required to run in realtime. It's time
might be slowed down enormously. I wasn't sure if you wanted to consider
that at all, but it seemed you might, so I brought it up.

> Each atom
>would need to be imitated with its own computer; given that we do not
>have a small number of atoms in our brain, and a full imitation requires
>not just the imitation of the our brain but also at the least the input
>and output to all our peripherals (otherwise known as eyes, ears, nose,
>hands ... remember what I said about input and output being constant
>while we remain alive) then the possibility of getting together enough
>atoms to do this imitation raises its ugly head. Moreover, we'd also
>have a highly parallel machine, with every computer imitating an atom.
>

No Thomas, as long as time is *not* a factor, you can do everything with one
computer, one Turing machine running on one, one-dimensional tape, just by
managing appropriate descriptions of everything you are modeling. That's the
beauty and power of computation. But again, this makes no claim that the
system could execute in anything approaching realtime.

>
...
>Furthermore, such a computer would also have to show the ability to
>grow and change, two features our brain shows. 

No, the computer would not have to grow or change. The growth and change
would be reflected in the pattern of descriptions it built up and updated.
Again, I'm not claiming a system like this could operate in realtime.

Mike Perry

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=15030