This is the mail archive of the
gsl-discuss@sourceware.cygnus.com
mailing list for the GSL project.
[this is the full note] the game's afoot, Watson
- To: GSL discussion list <gsl-discuss AT sourceware.cygnus dot com>
- Subject: [this is the full note] the game's afoot, Watson
- From: Mark Galassi <rosalia AT lanl.gov>
- Date: 17 Sep 1999 10:29:01 -0600
OOooops -- I hit C-c C-c instead of C-x C-x.
--
Amigos, I just got a call from Steve Adler, promoting my talk from a
"poster" to an "invited talk" at the Open Source/Open Science
conference.
This ups the ante: people might actually pay attention to the talk,
and I want to do a good job of explaining what we are doing, with the
goal of eventually convincing scientific program maintainers (gnuplot,
octave, ...) to use GSL.
Next week I will prepare a set of slides (probably simple bullet lists
in TeX -- they'll have a Linux box to project with gv), and it now
becomes more important for y'all to let me know what things you want
me to emphasize.
Just off the bat (unpolished), here are some points I am thinking of
making central to the 30 minute talk:
* Describe everyone's motivation coming in to GSL. Please send me
yours. Mine was that I developed a prototype of Gnudl, and realized
that I would need a complete scientific subroutine library to give
all those primitives. Other people wanted a high quality free
replacement of Numerical Recipes, or an infrastructure to provide to
all free science software. I will mention those motivations, and
please give me others you find important.
* Give an overview of the functionality covered, and of the authors
(let me know if you are an author and do not want me to talk about
who you are and where you work).
* Talk about the goals Jerry described: they are close to the original
motivations, but they have evolved somewhat.
* Talk about (pop music) programming language and infrastructure
issues. This involves VHLL wrappers, C extensions (error/assertion
support), C++.
* Mention the future: 0.5 release at around the time of the
conference; 1.0 release this winter; Jerry's thrust on linear
algebra and C++. Jerry, could you point me to standard material
describing the failure of linear algebra efforts to date?
* Mention who is using GSL and for what. I actually know little about
this, except that I sometimes hear people mention that they use it
quite happily in their work. Could y'all send me a note telling me
what you use GSL for?
This is quite crude; please let me know soon what you suggest for this
talk. It's on behalf of all the authors.