---------- Original Message ----------------------------------
From: "Steve Petermann" <steve@spetermann.org>
Date: Fri, 28 Nov 2003 20:25:13 -0600
>Glenn wrote:
>>>>>>>>>>
>> Or a process called simulated annealing, which is used in geophysics to
>design models of the subsurface. These things use chance based upon random
>number generators to mutate the model looking for a better fit to the real
>data. This type of procedure is quite widespread and shows that designers
>do use chance.
><<<<<<<<<<<
>
>I didn't say a designer couldn't use chance, just that no good designer
>would let chance alone dominate or control a design. Your example could be a
>good analogy of how a divine designer could use chance to create and solve
>problems. However, this is also a good example of *intelligent* designers
>using natural randomness to create. I think you would agree that chance
>alone does not solve these problems but the intelligent utilization of
>resource available, including chance.
>
>Steve Petermann
The term 'good designer' is a relative term. IN cases like I mentioned with trying to design an accoustical impedence volume from a seismic volume, the number of possible volumes is so large that humans can't examine them all in the age of the universe. Thus we use chance to look around the space of all possibilities to find a solution which is good. It isn't necessarily the best, but we have very little way of know a solution is the best even if we stumble upon it. Thus, the term good designer is one that comes from a high pedestal and implies that someone can over look the entire field of possible solutions and choose the best of all of them. This simply isn't possible with most realistic technological problems the world faces.
Indeed, the problem I am thinking of has more possible outcomes than contained in the human genome. Yet via random design and comparison (which mimics random mutation and selection) we can design a very useful solution to our problem. And I might add, one that matches reality quite nicely. Reality being defined by the number of increased barrels we get out of the ground.
Received on Sat Nov 29 09:49:33 2003
This archive was generated by hypermail 2.1.8 : Sat Nov 29 2003 - 09:49:34 EST