I have previously asked how why we should trust computer modeling of
dire predictions of global warming, and have been roundly chastised by
those who say: How dare you doubt us? as if questioning so-called
authority is out of style. One respondent said: A single outcome of
computer modeling is unscientific, to which I agreed. So, try this,
you computer modeling enthusiasts: Build a computer model which will
guarantee, repeat guarantee, a positive monetary return, no matter how
small, on placing winning horse racing bets in the U.S. This is hardly
a single outcome situation.  If you cannot do that, then shut up. If
you indeed have such a program, please share it with all of us. We
could use the money. And while you are at it, please build an
infallible computer model which guarantees that anthropogenic
greenhouse gas emissions are certain to cause disastrous effects on
global climate. We need that, and right now.

Hi, monroe22-ga:

If I have said anything to discourage your doubt, whether of authority
or facts, then I regret having phrased matters so poorly.

Best known for his contributions to the foundations of quantum theory,
Edwin Schroedinger wrote a book What is Life? (1944) which drew
attention to a distinction between coherent systems in physics and
chemistry, which produce "order from disorder" through statistical
behavior of large numbers of interactions, and those in biology, which
produce "order from order" through the critical operations of a
relatively small number of molecules which must faithfully transmit
their information down the generations:

[What is Life?]
http://plato.stanford.edu/entries/life/#3

It would be amazing if one could so easily produce certainty from
uncertainty as your Question asks of "computer modeling".  Since I
have no such program, it seems that all I can offer you is silence.

mathtalk-ga: I bow down before your expertise, and I do not say that flippantly.
I wish I had a fraction of your grasp of a difficult subject, and in
no manner do I dismiss your comments. I can only speak as one who has
thirty years of brutal industrial laboratory experience, plus more in
the field. Because of this, I am a certified cynic in all things
*scientific* which by no means proves me correct in disputing concepts
such as computer modeling. Yet, I am grateful to anyone who advances
my knowledge by unequivocally proving me wrong.
Best regards,
monroe22-ga

monroe22-ga,

Looks like me and mathtalk just can't stay away from your
questions...we can't ANSWER them, but we can't stay away either.

I'm curious to know...what do you think of weather forecasting?  

The weather models NEVER give a guarantee, but instead, provide
predictions with a 60% certainty of snow or 90% certainty of rain, or
whatever.  It's easy to dismiss these as mere guesswork (which they
unquestionably are).  But it happens to be pretty good guesswork, and
it's a skill that's improved dramatically since the days when I was a
mere lad, largely because the data inputs and the models have
improved.

N'est ce pas?

palafala-ga: First, I am proud of being able to spell your difficult
handle, and it would be marvelous if you, mathtalk, and I could
somehow meet and exchange thoughts. Secondly, if my questions are
provocative, then we all profit. As I said, please prove me wrong. I
need all the help I can get.
  Sure, good guesswork is better than nothing, but it really bothers
me that momentous policy decisions are contingent on computer models.
I wish I had retained the news item about astrophysics in which an
authority said a recent discovery ( I believe from the Hubble
telescope ) proved our computer models were dead wrong. That was some
years ago, and probably was the origin of my skepticism concerning
computer modeling. But much before that, I was personally involved in
a program which asked us to assign numerical weights to problematic
unknowns. That did it. So, to sum it up, I am all for guesswork. Each
of us uses it constantly in our daily lives. But please don't present
some *scientist* or *expert* guesswork as incontrovertible proof.

I don't think there's a single scientist in the whole history of human
endeavor who has ever proved anything!  Not Newton, not Einstein, not
even Bill Nye the Science Guy.

Science doesn't prove things.  Lawyers get to prove things like guilt
or innocence in courtrooms (but do they really?).  Mathtalk and his
colleagues can construct mathematical proofs (but do *they* really
prove anything), and theologians like to prove the existence of
you-know-who.  But science -- and scientists -- generally stay away
from the proving business.

So anyone who tells you that they have incontrovertible proof of
global warming is full of hooey.  But then again, anyone who claims to
have proof that the sun will rise on time tomorrow -- or that an apple
will fall to the ground if I let go of it -- is just as mistaken.

No proof.  None.  Zip.  It's all guesswork...but much of it holds us
in very good stead.

pafalafa-ga: I was proud of spelling your handle, and then found I am
wrong once again. Exactly what is the origin of your mysterious,
unspellable moniker?
  Anyway, I presume that your last comment implies that the
uncertainty principle trumps any and all statements of what we call
facts. That moves us into the realm of metaphysics, which is
irrelevant to my question, which is, why should anyone put any faith
in computer models? If you believe that computer models are
infallible, please say so. If, on the other hand, you hold that
computer models are  but educated guesses, why, I agree 100%.
Best regards,
monroe22-ga