Scientific Libraries in Python

[William Park]

Horatio Davis <horatio at qpsf.edu.au> wrote:
> On Sun, 11 Nov 2001, William Park wrote:
>> Simply including the Fortran/C libraries (and their wrappers) as part
>> of Python distribution wouldn't be a problem, other than licenses.
>> But, so far, nobody made that decision or see the need for it.
> 
> Not all people want to use Python for scientific computing. So,
> putting such specialized codes in the standard library would be
> pushing "batteries included" a little far.
> 
> What I am advocating is a standard extension library for doing
> scientific computing in Python, in the same way that Numeric is the
> de-facto standard extension library for doing number crunching in
> Python.
> 
> Of course, this thread has seen no comment yet from the authors of
> scipy and Scientific Python (which seem to be the two key codebases),
> or the other software projects which build on these.  There might be
> some perfectly obvious reason why this is a horribly bad idea...

Most of the time, I just want access to library routines, especially
higher functions, matrix system, and differential equations.  It would
allow me spend time working on the problem, not scavenging around to
find libraries.

-- 
William Park, Open Geometry Consulting, <opengeometry at yahoo.ca>.
8 CPU cluster, NAS, (Slackware) Linux, Python, LaTeX, Vim, Mutt, Tin