Gordon made some good observations about the credibility of models. I
fully agree that the mere existence of a model which could produce the
observed or postulated phenomena is not in itself a proof that the
phenomena occur, or occur according to the dynamics of the model. However,
a model can be investigated and compared to real data. A model that
survives such comparisons gains credibility -- the more tests it survives,
the more credibility it gains. Models like Kauffman's and Prigogine's and
Mandelbrot's and... have not survived many tests yet -- they haven't been
arund that long, but they seem capable of explaining phenomena people want
to explain, so they will be around for a while -- at least until they are
shown to have significant deficiencies.
Bill Hamilton | Vehicle Systems Research
GM R&D Center | Warren, MI 48090-9055
810 986 1474 (voice) | 810 986 3003 (FAX)