Re: Linus Torvalds quote

In message <o0aem3hn9nbs9jvlgnm1d4fiqv4ckgbfpe@xxxxxxx>, John Vreeland <vreejack@xxxxxxxxxxxx> writes
On Mon, 17 Dec 2007 21:00:35 +0100, nospam@xxxxxxxxxxxxxxxx (J. J.
Lodder) opined:

Paul Ciszek <nospam@xxxxxxxxxx> wrote:

In article <68a5399f-acd3-4f18-b531-77938f17c30e@xxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Kermit <unrestrained_hand@xxxxxxxxxxx> wrote:
>On Dec 17, 10:55 am, conrad <con...@xxxxxxxxxx> wrote:
>> On Dec 16, 9:33 pm, Kent <musquods...@xxxxxxxxx> wrote:
>>
>> > Don't EVER make the mistake that you can design something better than
>> > what you get from ruthless massively parallel trial-and-error with a
>> > feedback cycle. That's giving your intelligence _much_ too much
>> > credit. -- Linus Torvalds
>>
>> > Seems to me this describes evolution as well as free software
>> > development. What is evolution but a "ruthless massively parallel
>> > trial-and-error with a feedback cycle.
>>
>> > Kent
>>
>> It isn't parallel though. All organisms stem from a single
>> point. But it is a kind of trial-and-error or error-by-trial
>> system with a feedback cycle. The error-by-trial
>> being mutations and the feedback cycle being natural
>> selection.
>
>Not parallel? How about multiple organisms with not-quite-identical
>genomes being simultaneously tested?

And then you take the winners, mix up their genomes, and generate a large
number of new trials from them. The larger the gene pool (up to a point)
the faster evolution happens, according to a recent article I saw recently.
Sounds parallel to me.

Huh? The consensus is that -smaller- populations can evolve faster,
because an individual with an advantageous mutation
is -relatively- more important in the small population.
A huge gene pool needs more time to adopt it,

Jan

While a smaller population might appear to evolve faster due to rapid
fixation, that is not the same as adaptation. Fixation is much more
likely to happen solely due to drift in a small population. If you
are just looking at gene frequency changes you might be tempted to
call it fast evolution, but it has nothing to do with adaptation. A
smaller gene pool cannot maintain as much variation as a larger one.
Without a large library of possibly adaptive alleles to draw from in
times of environmental change a small population is much more likely
to fail to adapt.

A standard result of population genetics is that the rate of change due to neutral drift is independent of the population size - a neutral change is fixed faster in a smaller population, but the greater number of mutations occurring in the larger population cancels this out. (Both fixation rate and mutation rate are proportional to the population size.)

Things are not so simple when selection and varying population size are thrown into the mix.

of course, ianab

--
Alias Ernest Major

.

Relevant Pages

  • Re: Natural selection resolution and concurrency
    ... and stall evolution? ... likely time to fixation if it does fix. ... there are far fewer mutations in the small population so the ... luck and good mutations will get more opportunities (as will bad ...
    (talk.origins)
  • Re: Natural selection resolution and concurrency
    ... and stall evolution? ... likely time to fixation if it does fix. ... there are far fewer mutations in the small population so the ...
    (talk.origins)
  • Re: PE
    ... non-neutral appears when the selection coefficient 's' is close to ... copy number) and beneficial mutations growing to fixation ... the probability of fixation is 2s. ... the major factors in evolution. ...
    (talk.origins)
  • Re: The Theory of Evolution is Mathematically Irrational Round 2
    ... but it doesn't give significant benefit when evolving new alleles. ... 111 "Five point mutations in a particular Beta-lactamase allele ... conclude that much protein evolution will be similarly constrained. ... Fixation of neutral mutations does not occur in only two ...
    (talk.origins)
  • Re: Does anybody want to proofread and criticize a 7000 plus word
    ... Statistical Probability in the Evolution of Mendelian Populations ... advantageous mutations that frequently recur, ... and some elements of neutral molecular evolution. ...
    (talk.origins)