From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Sat Dec 08 2001 - 10:41:00 MST
"Robert J. Bradbury" wrote:
>
> But you can easily see how it might be advantageous to design
> more chromosomes (redundancy) and greater numbers of alleles
> into a population so it has more flexibility. Nature developed
> a system that is "good enough" but nowhere near the limits.
This needs to be phrased carefully, since it's not obvious what the
advantage is to the individual of carrying contingency alleles. How could
Nature design populations rather than individuals? Well, over very very
long periods of time, the species that exist will tend to be the
descendants of species that retained adaptive capability and genetic
flexibility. Even if, in the short term, a species has better
error-correcting methods or less variance with respect to a temporarily
invariant environmental condition, over sufficiently long periods of
evolutionary time that species will be outcompeted by other species, even
though as individuals they were temporarily superior. If this happens
through enough generations of species (not just generations of
individuals), there will be some lasting residue of "species fitness" and
"evolution of evolvability", even though locally species will still tend
to evolve for maximum individual fitness without regard to future
evolvability.
-- -- -- -- --
Eliezer S. Yudkowsky http://singinst.org/
Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:12:24 MST