Re: AAT Theory

H. M. Hubey (hubey@pegasus.montclair.edu)
2 Oct 1995 00:05:34 -0400

hubey@pegasus.montclair.edu (H. M. Hubey) writes:

>anyone can produce. That's the reason it gets mentioned all the
>time. As far as bone-eye-balling goes, it's in even a lower
>state of science than economics. The reason being that there are
>many more things that can be measured and made scientific than
>in physics but it doesn't get done. That's the reason for all the
^^^^^^^

This should have been economics. I'll explain further.

For physics simple measurements of time, space, mass etc
are sufficient to get going. ONe can derive velocity,
acceleration, force etc as part of the theory. The thing
is that measurements in paleontology fall far short of what
is needed. Simple measurements of bone thickness and length
are not enough. They don't even scratch the surface. We are
mostly interested in shapes and relative sizes and proportions.

Ratios of lengths, and other things are a good start but even
more important are measurements that have to refer to shape
and form. See below.

>The first good start in this direction was D'arcy Thompson. We have
>even better methods now but I don't know anyone who's doing any
>work in this area. It looks like it's all talk, so far. And you
>are one of the biggest perpetrators of heat/flame on this newsgroup,
>as far as I can see.

I think conformal mapping would have been very useful thing to
follow up on in D'Arcy Thompson's book. I assume that someone
must have done this already since 1927. There we'd have some
numbers to play with and these numbers can be used to create
relationships, functions, formulas etc.

These days with powerful graphics computers available cheaply
it should be an everday thing to try conformal maps on shapes
to see how one can deform one shape into another continuously.

Even more to the point, some of the morphing techniques used
in scifi movies are available on home computers and that
will carry the ideas even further.

Naturally some background in statistical inference or
correlation-regression analysis should also be included,
as it is done now in economics.

Lacking all this, the whole thing is still verbal. If I were
really cynical, I'd say pre-scientific, but I know how complex
it is thus it would be unfair to make that claim.

But it's time to climb down from the high chair.

-- 





				Regards, Mark

				http://www.smns.montclair.edu/~hubey