Re: Chaos and topographic diagrams

[Myron Tribus <mtribus@home.com>]

In recent posts Bob Adsett and Alla Linetsky have commented on the 
difficulty of using statistical methods to determine the efficacy of 
a treatment.  These comments and others like them remind me of a 
similar problem with which I dealt about 30 years ago.  At that time 
the question was the efficacy of cloud seeding to reduce the dangers 
from hurricanes or to modify the weather in other ways.  There were 
many debates over how to decide if cloud seeding worked or not. 
Weather systems are very complex and methods of measurement are not 
all that simple.

My analysis was printed in Science, April 10, 1970, pg. 1 under the 
title, "The Physical View of Cloud Seeding".  The thrust of my 
argument was that statistical methods were not useful and were 
actually harmful.  My point was that they kept us from developing a 
greater understanding.  The paper was given in Santa Barbara at a 
meeting of the American Meteorological Society and since there were 
several prominent statisticians in the audience, we had a marvelous 
debate.  No minds were changed, of course.

The arguments are too long to repeat here.

Myron Tribus,  350 Britto Terrace,  Fremont, CA 94539
Ph:510 651 3641  Fax: 510 656 9875   e-mail: mtribus@home.com
People need to be taught how to listen. No one needs to be taught how 
to interrupt. (Vygotsky)