X-Message-Number: 8217
From:  (Thomas Donaldson)
Subject: Re: CryoNet #8200 - #8202
Date: Sat, 17 May 1997 21:40:21 -0700 (PDT)

Hi again!

I've decided to make the self-referential problem a little clearer (I hope).
Clearly if a computer is simulating the universe it must also simulate
itself. If it simulates with a delay, then it might work out just what the
results of its simulation will be, and report them. This would force events
in the simulated world to take place with a delay.

But that does not solve the problem. We'd all want to know the results of
the simulation, and our activities would depend on them. Therefore its 
simulation will affect the calculations it does to find out its simulation...
after all, we are in the Universe, and we can act on it. It cannot simply
wall itself away from the rest of the Universe. Its results will affect our
behavior, and therefore the behavior of the Universe (as a whole).

Of course, if it were NOT to report the results of its simulation, then it
has a similar status to that of God: unprovable existence, unprovable value.
We are in the same kind of place as Perry Metzger: whether or not we are
simulations becomes irrelevant and unprovable.

The computer somehow must know the results of its simulation before it can
calculate them. That is no mean trick!

We could also suppose that simulation here involved no program other than
one which read off a tape giving the entire history of the Universe, including
the results of its reading the tape. This would solve the self-reference
problem, but at the cost of adding on this special reader to a history of 
the universe which will go on regardless. If that is simulation it's a 
rather odd version. Again, we might suppose that its results by some magic
cannot influence us at all: it reports them to the air, not to us. We're
back again to metaphysics: an unknowable computer which goes off and 
simulates everything. Why anyone would want to make such an object I do 
not know. Since it will be composed of matter, no doubt the best thing to
do would be to use the matter for something more useful.

Incidentally, for Mr. Clark: I do a lot of programming, too. Whether or not
this gives me some special status I don't know.

			Long long life,

				Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8217