X-Message-Number: 11253
Date: Wed, 10 Feb 1999 22:13:04 -0700
From: Mike Perry <>
Subject: Re: CryoNet #11242

Thomas Donaldson writes,

>Frankly I believe that the issue of funding more research into
>preservation of brains is worth far more than any discussion of whether or
>not we are finite-state machines. If we find out how to reversibly 
>cryopreserve brains, all these issues will probably have had far more
>discussion and possibly even agreement than they have now.
>

I won't argue this. As I recall, it was Fred Chamberlain who said,
many years ago, that worrying about the distant future and the larger
picture, with the immediate problem we face (mortality) is like
worrying about old age when someone has just fired a bullet at you
from 100 yards away. BUT, I still think the other, philosophical issues
have importance, and that we don't always have to choose one kind
of problem to focus on to the exclusion of the other. (In the case of
the bullet, we pretty much have to address THAT problem right
this instant, and don't have time at all for something else!) When both 
can be addressed without undue interference, why not do both? 

>Since 21st Century Medicine is doing such research, we need ways to 
>provide more funding for 21st Century Medicine, directed not at the 
>preservation of kidneys but at the preservation of brains. I would like
>to hear more from Saul Kent on what is possible here.
>
So would I.

>HOWEVER, for Mike Perry, here goes: first of all, although I may not have
>said so clearly, even a finite machine can go through an infinity of
>states.

"Finite" is not the same as "finite state". An analog device could be finite
but capable of an infinity of states. But I don't think the real world works
that way, i.e. a true analog device in this sense is impossible, 
though it can be approximated. Quantization imposes graininess,
even if space and time form a continuum. Instead, by reasonable
 indications the whole visible universe is, at the present moment 
(i.e. from our vantage point here and now), both finite and finite 
state. I mentioned the estimate of 10^98 bits
 for the information content of the present universe (p. 662, *The 
Anthropic Cosmological Principle* by Barrow and Tipler). There
is another, larger estimate that Moravec gives, of
10^122 bits (cited in Broderick, *The Spike*, p. 243). 
Moravec imagines you have converted all matter to low-energy
photons and that you can store 1 bit per photon, which means 
that even one atom of matter could store many, many bits (which
seems unlikely to me but I haven't seen it disproved). 
Anyway, still very finite, both in overall size *and* in number 
of possible, significantly different configurations, thus finite-state 
as well. But the universe is expanding, and even accelerating, 
and this makes it hopeful that its information content and number 
of possible states are increasing without limit. 
...
>You may wish to define a "finite-state machine" as simply a machine which
>is finite at any fixed time. It seems to me that this destroys the whole
>purpose of such an idea. Clearly such a machine need never return to 
>any previous state, nor have any other attributes belonging to a previous
>state. 


No, I would not define a finite state machine in this way, and in characterizing
a certain device such as a person or our universe as a "finite state machine"
I hope the reader will always keep in mind that I do not mean to say that, over

an infinite interval of time, this characterization is to apply. In the same way
a Turing machine is not a finite state machine--but it does become equivalent 
to one over a finite interval of time. Saying so
(and giving more details) will help some to better understand what sort of 
device a Turing machine is. Similarly I think that characterizing people as 
finite state machines, if the qualifications are reasonably understood, will 
help in understanding something about the nature of personhood.

Mike Perry

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=11253