RE: [boost] ffts & special functions

From: Ken Appleby (ken_at_[hidden])
Date: 2002-09-20 07:32:30

> From: Jason D Schmidt <jd.schmidt_at_[hidden]>
> ...
> I have some numerical code that might be in shape enough to think about
> having it reviewed.  Before I submit anything, I'd like to get an idea
> what people are looking for.  Here's what I have right now:

Most users of numerical libraries will already have made their own
provisions, so I think the main value that such boost libraries should add
is genericity and quality of interface.  And the simpler the functionality a
library provides the more valuable it would be because more people would use
it.  Perhaps libraries of mathematical and physical constants would have the
greatest value to effort ratio?  A library of basic statistical methods
would also be very useful.

However, I would be wary of code that is based on Numerical Methods for
reasons quite apart from the copyright issue.  For example, the fast random
number generator "An Even Quicker Generator" described in section 7.1
returns alternate even and odd numbers for all seeds.  NR claims it has
undergone "extensive tests."  See http://math.jpl.nasa.gov/nr/ for other
warnings.

Ken Appleby