--- Raymond Toy <toy@rtp.ericsson.se> wrote:
> >>>>> "Robert" == Robert Dodier <robert_dodier@yahoo.com> writes:
>
> Robert> no clear statement of terms of use in the
> Robert> non-SLATEC version. However, there is a statement
> Robert> in SLATEC which says SLATEC is in the public domain.
>
> I'll use the slatec version then. I suspect there's not much
> difference between them.
As Brian Gough points out, there is a version (GPL) of Quadpack
in the GSL which contains bug fixes. That version is preferable,
although it is in C which may be harder to translate to Lisp
than Fortran; perhaps GSL Quadpack can be called through FFI.
FWIW I looked closely at the difference between the Netlib
SLATEC and Netlib non-SLATEC versions, and, ignoring comments
and whitespace, etc., the differences appear to be only these:
when Quadpack was brought into SLATEC, some clean up was
done, one bug was fixed (lower limit of integration in
dqawfe.f and qawfe.f), one was introduced (defabs vs resabs
in dqagpe.f and qagpe.f), and some decimals were chopped
off in the Gauss-Kronrod parameters.
See also the Quadpack bug list cited by Brian Gough.
For what it's worth,
Robert Dodier
__________________________________
Do you Yahoo!?
Free Pop-Up Blocker - Get it now
http://companion.yahoo.com/