Re: Duane Gish in Peru

In Glenn Morton's post, it was quoted:

> "Actually, The Design Inference contains little of genuine mathematics,
> but is full of mathematism, that is, the use of mathematical symbolism as
> embellishment, often only to create an impression of a scientific rigor of the
> discourse."
>
>

This is probably a bit harsh. There were two main
problems I saw with The Design Inference.

(1) You still need to find a way to assign correct
probabilities to the system you are considering
before you can apply the model. Winning hands in
a fair card game can be estimated accurately. Estimating
the likelihood that a sequence of signals from outerspace
are not noise can also be assessed with additional
information. It's application to evolution is dubious
because we simply don't know the odds for a lot of
things.

(2) Given you have a correct description of the probability,
you can then assess whether something was designed by setting
a lower bound on the probability. That boundary, although
reasonable, is not really a rigorous proof. So winning
10 random drawings in a row is clearly suspicious, but where
to draw the bar still seems to be based upon opinion.

So it does examine some very important questions,
however it is not so strong at answering any of them.
Received on Sun Oct 17 11:10:23 2004