>Real computers, though, present an extra intellectual challenge for us.
>I agree that computers are real things that do work and dissipate.
>Flip-flops flip, heat is dissipated from current flowing through resistances,
>etc. The interesting thing is that computers have been cleverly designed
>so that they implement the conceptual operation of a formal system.
This still does not satisfy me. I have to sideline with Penrose on this one.
Suppose we create a gigantic computer where each computational cell is a
"Chinese room". That is, there is a mailbox into the room (input) and a mailbox
out of the room (output). Inside this closed room there is a Chinese boy who
has some strict instructions (algorithm) of how to deal with the incoming data.
He then takes the data coming in and do with it what he is instructed to do and
puts his result in the other mailbox.
Now, if dissipation is all that is needed for a macroscopic
conscious/sentient unity to emerge then we must accept that both a computer
(either built from transistors or chinese rooms) can be conscious. But is it
really reasonable to assume that the work done by the boy somehow mysteriously
translates into some higher level unity? It's possible, but we are certainly
on very thin ice here. I personally have my doubts. I am positive that friction
(hence energy dissipation) is crucial to the process of sentience, but I am not
sure it is enough. What makes dissipative structures candidates for large scale
sentience is the fact that noise from as low as the quantum mechanic level is
blown up all the way to the macroscopic level. The dissipative structure is not
just a dissipating system, the dissipation actually leads to large scale
resonances. Here the dissipation is in direct contact with this resonance which
is not the case in computer systems where the resonance completely lies in the
domain of a software Gestalt. I may be old fashioned but I feel sentience is
rooted in the Real forces.
Onar.