Hi,
I'll take a stab at answering this.
In comparison with Matlab, perhaps the biggest thing that Sage is
missing is the huge 3rd-party ecosystem of scientific computing and
engineering scripts/plugins and the extra toolboxes for specific
applications. Sage's use of python helps a lot with this for some
things, but not others. As an example, I've been trying to get a
friend of mine to try Sage, but he isn't convinced it can help him. A
few weeks ago he obtained a scientific instrument that dumped out data
in its own strange, proprietary format and he needed something to
convert it. In 5 minutes he could find a Matlab script that did
exactly what he wanted. This is a chicken-and-egg sort of problem -
we need scientists and engineers to use Sage, but they won't try it
until lots of scientists and engineers use Sage. Still, I think the
use of python helps a lot and I think we have a chance to make some
inroads there.
Compared to both Matlab and Mathematica, another big weakness is
plotting, both 2D and 3D. But we're not too far off, and with
matplotlib/tachyon/jmol one can do a lot. Sage just needs some
serious polishing of what it has. I hope to do a little on that with
the 3D stuff. It is worth remembering that the jmol inclusion is
still only about 6 months old.
Mathematica has some other areas in which its clearly better. One of
them, like Matlab, is the amount of 3rd party code. It is also
currently better at symbolic manipulation, things like substitutions
with rules. There is current work on Sage's symbolics (by Gary
Furnish and I'm not sure who else) that will hopefully improve the
situation there. I think we should make it easier to use regular
expressions too, but I'm not volunteering for that :) Finally,
Mathematica's Manipulate command is still better in many ways than
Sage's @interact. I am very hopeful for the future there, since
@interact is quite new and it does almost all of what I really need.
I am trying to learn enough javascript to help extend it, and I think
within the next year we'll catch up more.
One thing to keep in mind is that so far Sage has moved very fast in
catching up. Mathematica is 20 years old, and didn't get its
Manipulate command until last year. Matlab is 30 years old, and my
impression (I haven't used it in years) is that it is still fairly
weak in abstract mathematics.
Cheers,
M. Hampton
On Jun 9, 2:50 am, "Jurgis Pralgauskis" <
jurgis.pralgaus...@gmail.com>
wrote:
> I am making a presentation on SAGE in university (for PhD students
> math/informatics)
> so this is quite general quoestion
> in university context I can stress first years course
> and for research I understand that it depends :)
> but well, people working with the soft should feel some points
>
>
>
> On Mon, Jun 9, 2008 at 1:41 AM, William Stein <
wst...@gmail.com> wrote:
>
> > On Sun, Jun 8, 2008 at 2:25 PM, Jurgis Pralgauskis
> > <
jurgis.pralgaus...@gmail.com> wrote:
>
> >> Hello,
>
> >> as the vision of SAGE is to be alternative to MMM,
>
> > True.
>
> >> what do You think is mostly lacking now?
>
> > This depends a huge amount on what you want and why
> > you are asking.
>
> >> and, what is better than in MMM ones (except open source ;)?
>
> > Again, this depends a lot on what you want and why you
> > are asking. All these programs we're discussing are massive.
>
> >> maybe there's a wiki page likehttp://
cmsmatrix.org?
> >>
http://web.bryant.edu/~bblais/bryant/numerical_computing/python_matla...