Hal Finney wrote:
> 
> My one concern is that if Wolfram is right and our universe is a random
> program from some set, and if there are much more than on the order
> of 100 bits in the program, we will never be able to find the right
> program.  If the nature of the program space is similar to what Wolfram's
> explorations suggest, that most of the space is unstructured and there
> is no way to identify the likely fruitful programs, there will be no
> way for us to know if we are on the right track or not.  We won't have
> the hope of finding a program that "almost works" and then successively
> refining it to get closer and closer, because the true program will be
> completely different from an almost-true program.
> 
> The situation will be something like the search for a cryptographic key,
> where you can't really hope to get closer and closer until you get it.
> You're stabbing totally in the dark until you fall upon the right one.
> And if the search space is too large, you will never find the answer.
> 
> Hal Finney
> 

This is exactly the problem I have with the idea that we live within
one particular computational history, for all time, but can never know
which.

There will be constraints on the set of possible histories in which
can be lived, which can be uncovered and linked to a theory of
conscious observation. All else is pure contingency. So other ideas
proposed on this list - that we can be identified with a whole sheaf
of computations that are continually diverging - seem far more appropriate.

----------------------------------------------------------------------------
A/Prof Russell Standish                  Director
High Performance Computing Support Unit, Phone 9385 6967, 8308 3119 (mobile)
UNSW SYDNEY 2052                         Fax   9385 6965, 0425 253119 (")
Australia                                [EMAIL PROTECTED]             
Room 2075, Red Centre                    http://parallel.hpc.unsw.edu.au/rks
            International prefix  +612, Interstate prefix 02
----------------------------------------------------------------------------

Reply via email to