Re: Challenge for Ken Shackleton

On 20 Apr 2006 13:30:46 -0700, "Seanpit"
<seanpitnospam@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:

<snip a lot of arguments that have been thrashed over so many times
already>

I think I will jump in, to rescue Ken from loneliness (as he put it in
another thread).

The problem is that you (Seanpit) do exactly what I complained about
in that other thread (Bacterial Evolution question): you use notions
of information simplistically without regard to the biological
context.

Biological evolution is not simply one sequence of letters randomly
mutating into another sequence or even serially replacing individual
letters with alternative letters. There are well known and well
studied mechanisms of gene duplication of translocation and copying so
that functional bits and pieces, whether whole protein families or
amino acid sequence motifs get cut and pasted together. There is also
a tremendous amount of redundancy in the way biology works so that
non-functional intermediate steps are not necessarily deleterious. In
other words, looking at "sequence space" and trying to find "islands"
of functionality and measuring probabilities of stepping from one
island to another is completely fallacious -- that is not the way
cells work. Biology has mechanisms to hop across gaps in the sequence
space "preferentially finding safe spots to land" (note carefully: I
am speaking figuratively here, not imputing teleological mechanisms at
work). Your "exponentially decreasing probabilities" are simply
wrong. You fail to understand that large scale changes can be
induced by gradual small steps leading to a rather significant
re-organization of the entire control system leading to a large jump
in result. You fail to understand that we expect to see only
relatively small evolutionary changes using the techniques and the
time span of observation that are available to us. You fail to
understand that the large evolutionary changes that we talk about have
left their unmistakable imprint on the genome, the biochemistry, the
developmental patterns, the morphology, and the ecology of living
things. You fail to understand a lot of things.


.

Relevant Pages

  • Re: Choice question; Was: metric spaces and convergent sequences
    ... R can fail to be Frechet, i.e. not every accumulation point x ... of a subset A may be reachable by a sequence in A. ... All Lindelof subspaces of R may be compact. ...
    (sci.math)
  • Re: DNA vs. AI
    ... sequence databases is growing at a rate outstripping Moore's law. ... Computer Science and Computational Biology". ... Proteins and Nucleic Acids" is over 350 pages. ... The field is going bananas. ...
    (comp.ai.philosophy)
  • Re: Cantor Confusion
    ... I say in this case because not every sequence has a limit. ... Factoring decimals is a muggs game. ... That one can construct a rule that fails does not mean all rules fail. ... If the nth digit of the nth number in the list is not 2, ...
    (sci.math)
  • Re: Getting Column Titles
    ... Jeff ... "John" wrote: ... >> I have tried the code you supplied, but have one question, why does it fail ... > sequence so those will have to be addressed. ...
    (microsoft.public.project.vba)
  • Re: Anthropomorphisms and the inherent periodicity of the DFT
    ... i fail to see this as a misuse of the language to make a point and to ... There is nothing that makes the periodic extension principle more fundamental than the filter bank perspective,the orthogonal decomposition perspective, or the sampling of the Z transform on the unit circle perspective. ... If a signal is actually periodic and matched to the sample rate so that the discrete sequence repeats itself, periodical extension of a single period is fine with me. ...
    (comp.dsp)