Eliezer commenting on my genome improvement strategy commented:
> This needs to be phrased carefully, since it's not obvious what the
> advantage is to the individual of carrying contingency alleles.
Yes, but we are *already* carrying around (and paying associated
costs) for the 90+% junk DNA. (The only caveat being if the
junk DNA functions as a free radical sink -- then it is actually
serving a purpose).
> 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.
That was my point about having chromosome pairs. It seems probable
that pairs combined with multiple genetic loci functioning in an
additive fashion is much easier than dealing with triplet or quadruplet
chromosome copies. Chromosome pairs and allelic variation have
provided enough "adaptive capability" that species can evolve
and survive. But that in no way suggests that the genetic memory,
adaptability or survival capacities are maximized by the specific
configuration most species on the planet have.
> 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.
No argument. The substrate (DNA repair, sex, chromosome pairs, etc.) evolved
to the point where species survival and individual fitness were sufficient
to allow life to both survive and evolve. That in no way suggests that
we are currently on a maximum extropy vector at this time. A maximum
extropy vector requires looking at the environment and determining
a survivability or information content quotient (these may be orthogonal).
It requires intelligence to predict future environments (one has to understand
both laws of physics and the evolution of the universe). That determines
the present value of current information and therefore what should be
preserved for future usefullness. An even more intelligent valuation
function determines the cost of recreating the information and when
precisely it might be needed. If recreation is cheap and it will not
be needed in near-time, then discarding it is reasonable. Nature discards
information without consideration of its future usefullness.
I have no problem with Nature having built-in a "foundation" of fitness
for evolution -- but the hazard function would appear to limit our
"evolution of evolvability". If the hazard function is decreasing
(Jupiter removing comets) then the pressure to evolve foundations
that promote the development of species that have enhanced evolvability
or survivability seems to be diminishing. There may be a planet in
the galaxy with no Jupiter, around a larger star (higher UV flux,
greater energy, faster evolution) in comparison to which we look
like slow-pokes. It is an interesting fundamental question --
what is the substrate that can support the highest rate of evolution
while preserving the greatest amount of previously generated information.
Robert
This archive was generated by hypermail 2b30 : Sat May 11 2002 - 17:44:24 MDT