Re: GSL API design flaws

["Gregory R. Warnes" <warnes at biostat dot washington dot edu>]

On Mon, 17 Jul 2000, Gerard Jungman wrote:

  GJ>> "E. Robert Tisdale" wrote:
  GJ>> > 
  GJ>> > Brian Gough wrote:
  GJ>> >
  GJ>> > > The goal of the project is really just to provide a free library
  GJ>> > > equivalent to existing proprietary libraries
  GJ>> > > such as IMSL, NAG and Numerical Recipes.
  GJ>> > > So we just follow the existing conventions
  GJ>> > > from these libraries with some minor improvements.
  GJ>> > 
  GJ>> > That's a disappointment.
  GJ>> 
  GJ>> Yes, it is. But we're not the source of the
  GJ>> disappointment; it is a simple fact that the
  GJ>> state of the art "products" in this field
  GJ>> fall far short of technical expectations.
  [...]
  GJ>> GSL is not the place for free jazz explorations
  GJ>> on numerical library themes. It's just a hammer.
  GJ>> You use it to pound nails. Higher levels of
  GJ>> sophistication must wait. I don't like it any
  GJ>> more than you do, but those are the facts.
 
It may be dissappointing that GSL isn't the next generation tool, but it
is very useful as it is.  

In addition, the availability of a free software numerical library that is
complete, reliable, and well maintained will provide a nice foundation for
work on future numerical libraries.

-Greg