Re: supernatural observation and faith def.

David J. Tyler (D.Tyler@mmu.ac.uk)
Wed, 9 Oct 1996 15:13:37 GMT

Further comments on Darwinian design techniques.

An engineer who has been following this thread has sent me the
following. It is a good example of the "real worl" applications of
these techniques. The person concerned has given me permission to
post this.

----------
I have used GA's some in my work -- for designing dampers that use
"smart fluids" -- and while GA's definitely are good at finding
optimum designs in some problems, I would quibble whether they are
designing anything. A better description is that they are
discovering a design by trial and error. I suppose you could say the
same of a human designer, but the process is different. The human
designer makes intelligent guesses about what to try next, while the
GA just tries things. To solve a problem using a GA you frequently
have to play around with parameters that determine mutation rates,
crossover rules, etc., and you may have to try different fitness
measures. The people who hype GA's never tell you about that. I
don't rule out evolution as one of the tools that God uses to make
His creation more robust and capable of adapting to external
disturbances. But the idea that a process like evolution could
explain the complexity of life is preposterous.
-------

Best wishes,

*** From David J. Tyler, CDT Department, Hollings Faculty,
Manchester Metropolitan University, UK.
Telephone: 0161-247-2636 ***