The measure of programs with high precision would probably be
about the same as that of those with low precision, unless for some
reason errors would blow up in which case high precision would be
selected for.
Actually there might be an interesting question here. Using a UD
to get higher and higher precision is fine - unless the program itself
needs to single out one brain state and halt, in which case the chosen
precision would have to be some fixed (for each program) number. But I
doubt that could be used to experimentally distinguish the
possibilities. Another possibility is that the program would just need to
either do as above or single out a convergent series of brain states and
keep printing them to the same region, in which case this remark does not
apply.
Marchal wrote:
>(°) I agree with Jacques Mallah that the measure depends on the number of
>implementations
Please explain that remark.
- - - - - - -
Jacques Mallah (jqm1584.domain.name.hidden)
Physicist / Many Worlder / Devil's Advocate
"I know what no one else knows" - 'Runaway Train', Soul Asylum
My URL:
http://pages.nyu.edu/~jqm1584/
Received on Thu Jan 06 2000 - 14:18:18 PST