Wade Schuette (941001) (schuette@eecs.umich.edu) --
[direct send with copy to CSG-L]

Hi, Wade --
Thanks much for the note concerning Tierra. I happened to have a big FAQ
on GA stuff, and found the email reference, which I looked at.
Unfortunately I can't download the stuff: I have only 1.5 MB storage in
my account (I dial up Fort Lewis College from home), and the Tierra file
is 1.9 MB long. Bottleneck. Also, the text references seem to be in TeX
format, which I can't handle. Anyway, I program in turbo C 2.0, and who
knows how compatible that would be with the source code?

I got into this because of wondering what would happen if a control
system modeled after the E. coli method of locomotion were superimposed
on some standard evolutionary model. The so-called E. coli method uses a
control system that senses some critical variable, compares it with a
fixed reference level, and converts the error into an increase or
decrease in the time between "tumbles" (mutations, when converted to
terms of a specific evolutionary example). This produces a very strong
bias in the behavior of the system without requiring that the
"mutations" themselves be biased. I'm trying to set up a very simple
organism in which random mutations can happen, to see what happens with
and without the control system.

I'm sure you can tell that I'm not an expert in this field. I'm not
trying to re-invent GA, but just to have a look at a particular
phenomenon, in my own peculiar way. I'll try to make my struggles public
so that Martin Taylor and others can help with problems I create for
myself. Your comments and suggestions would be _very_ welcome -- don't
be shy about horning in, even to say it's all been done before and I can
quit worrying about it. If people are already working on things like
this I sure don't want to fumble along in their dust, and there are
several other things I really should be doing instead. Actually, if it
turns out that there is anything novel in my idea, I would like to dump
the whole thing in the lap of somebody who would know what to do with
it, because I spend too much time out of my depth anyway.

Thanks again for your comments; hope to hear more from you.



Bill Powers