Re: tree-climbing hominids
H. M. Hubey (hubey@pegasus.montclair.edu)
12 Oct 1995 21:36:29 -0400
David Froehlich <eohippus@moe.cc.utexas.edu> writes:
>Let me make a statistical that you might actually understand. What you
>have just implied is the same as saying "I have flipped this coin and it
>has come up heads three times in a row. Therefore, I can predict the
>next coin flip"
>Pure, unadulterated bullshit!!!!
Froelich, you either don't read carefully or can't read. I already
gave a very precise and clear description of the mental model
I have of evolution. It has a deterministic component (like a DE)
and a random component (it could be forcing or randomness in
the coefficients). Stochastic processes which have deterministic
components have correlated spectra and this can be seen in the
autocorrelation of the sample functions.
Your simplistic example of tossing a coin might be good enough
for you but not for me. I encourage you to post more stuff like
this :-).. After a while even you will be able to see that you
are off. Coin tossing is uncorrelated.
>extrapolate in both directions. However, your basic misconception here
>is that evolution bears any resemblance to a vector. If anything it is a
>stochastichally random process mediated by selection. That selection may
Let's look at it at the genetic/DNA level. The space is n-dimensional
and obviously n is quite large. Each animal is represented as a
point in this n-space. That's a vector. The evolution of the point
or actually a whole cloud of points (since mutation requires
that we treat it as a stochastic process) is the dynamics of
the process. Simplifying it (as in the continuum mechanics approximations
of fluids or solids) we can reduce the dimensions and turn it into
a continuous variable stochastic process (exactly as I described above
and a couple of weeks ago).
It is a stochastic process, and I already said it. But it doesn't mean
that it has nothing to do with vectors. Your idea of a vector is
limited in conception, but you're getting closer and closer to understanding
what is going on.
>selective pressures are not static and consequently you can not
>extrapolate. Evolution is not a nice neat mathmatical equation that you
>can solve for. Remember that niche space is N-dimensional, probably more
>dimensions that we even realize.
If I knew the equations, I'd get a Nobel prize. That doesn't mean
that the model is any worse than the simpler models (many of them
incorrect verbiage). I think in fact that this is more correct.
There are three ways in which randomness enters into DE's
1) random ICs
2) random forcing/source terms
3) random coefficients
In practice all three are present in the real world phenomena
which we study.
>It might help if you actually learned something about evolutionary theory
>before you try to converse with it. Evolutionary scientists do not try
>to predict future events because the variables are not constant.
>Selection is not static (just look at the changing ideas about beauty in
>US society).
YOu really have to give up in this self-abuse. Do you realize that
there are people here who know more math and physics than you do
and can see what you are doing? You're making a fool of yourself
in front of thousands of people. I never said variables are constant.
YOu probably are trying to say that the coefficients (of what?)
are not constant. Yep. You are most likely correct if that's
what you mean. But that is one of the things that makes the
problem very complicated, isn't it? I already said that, didn't I?
Calm down.
--
Regards, Mark
http://www.smns.montclair.edu/~hubey
|