Symbolic computation on Fedora

[Matthew Saltzman]

On Sun, 2012-09-16 at 09:13 -0400, Suvayu Ali wrote: 
> Hello everyone,
> 
> First, thank you for your thoughts and suggestions.
> 
> On Sat, Sep 15, 2012 at 01:14:42PM -0400, Matthew Miller wrote:
> > On Sat, Sep 15, 2012 at 09:59:40AM -0500, Richard Shaw wrote:
> > > Ick... This thing (sage) is a mess of bundled libraries you'll
> > > definitely need to manage your own install. This looks to be a ton of
> > > work to unbundle all the libraries if someone wanted to package it
> > > officially.
> > 
> > There was some work on packaging it for Fedora a few years ago; not sure
> > offhand where that went. A fundamental issue, as you've noticed, is that
> > Sage is more a software *distribution* itself than it is an appliction.
> > This doesn't lend itself well to packaging.
> >
> 
> At the moment I am short on time, so cannot experiment with sage.  That
> said, Fedora has a long standing packaging effort for SAGE[1].  It seems
> to me sage needs a TeXLive-like distribution model to be not so
> intimidating for new users.  This however does not make the effort
> required to include it in Linux distribution respositories any simpler.
> 
> I'll try compiling my own in a few weeks when I have some more time and
> report back.

I have built Sage on Fedora 17 and RHEL 6 with no difficulty.  It is a
big ugly package, but it is self contained.  It would be nice to have a
package setup like TeXLive, but I have seen the discussions about the
challenges involved.

> 
> Cheers,
> 
> 
> Footnotes:
> 
> [1] <https://fedoraproject.org/wiki/SIGs/SciTech/SAGE>
> 
> 

-- 
Matthew Saltzman
Clemson University Math Sciences
mjs AT clemson DOT edu