Re: Nature article on ID

From: Iain Strachan <igd.strachan@gmail.com>
Date: Sun May 01 2005 - 05:05:59 EDT

> >
> It appears to me that all you attibute to ID is implicit in TE.

That was precisely the point I was trying to make. TE = ID
Evolutionism. God "shaped" evolution by designing the laws of the
universe so that the result was predestined. (As in the discussion of
the Sierpinski gasket).

Seems I've become a proper heretic now .. but I will say that it is
via the valid objections raised by ID arguments of, e.g. Michael Behe,
that I'm able to come to this conclusion. Anyone who has worked in
optimization (my own area was in training of neural networks and I've
also dabbled with genetic algorithms ) knows you have to design the
problem representation so it can succeed. In principle, certain
types of neural network can "learn" to reproduce any mathematical
function from empirical data, but if you don't apply sensible
pre-processing to the data (scaling, transforms etc) then the neural
net doesn't stand a chance of learning. (Even with much more powerful
optimization techniques than Genetic Algorithms, like Quasi-Newton
optimization). By the same token, Genetic Algorithms aren't a black
box that will solve any problem you want - to get them to work you
have to design the problem formulation so they have an easy time
solving it because they can't solve difficult problems where there are
lots of "cliffs" requiring several simultaneous changes.

Just how God could have designed a universe where there weren't these
"cliffs" in fitness space is quite beyond me, and I suspect anyone at
the moment, but to doubt that He could have is surely to doubt his
omniscience.

Iain.
Received on Sun May 1 05:08:30 2005

This archive was generated by hypermail 2.1.8 : Sun May 01 2005 - 05:08:31 EDT