Re: [Axiom-developer] Re: axiom for windows

[C Y]

--- root <address@hidden> wrote:

> > To begin:
> > 
> > I downloaded the current windows version of axiom. The one clearly 
> > highlighted as "recommended." Although TexMacs is mentioned 
> > prominently,  nowhere does it state that TexMacs is a separate
> > program  that the user must download independently of axiom.  Why 
> > not come out and say it, instead of assuming the user will infer 
> > that from the discussion or be sufficiently knowledgeable of the 
> > open source world to understand how the pieces interact with each 
> > other?
> 
> Ah, well, that's actually been a point of discussion. I'm in favor of
> packaging TeXmacs with the install script but the developers are not
> all of the same mindset about this. I've forwarded this comment to 
> the developers list and we'll address it with documentation and
> possibly combining TeXmacs into the install.

I'm still not quite sure why we can't have our cake and eat it to on
this - provide a non-TeXmacs installer for people who don't need it,
prefer to install their own TeXmacs, or think it's too big, and provide
a larger "kitchen sink" installer for the more typical user.  If I
understand correctly, the final exe install files can be hosted just on
the savannah servers, correct Tim?  Pounding your server to fine power
is definitely NOT one of the goals ;-).

> > I tried some of the CAS stuff, and it worked nicely - lovely
> > looking output. I'm not sure the strong typing is something that 
> > beginning students can handle, but that is an educational 
> > issue.              
>  
> Yes, it's certainly a difference between axiom and the other systems
> like maple and mathematica. Axiom is much more tailored to the 
> computational mathematician. Maple and Mathematica are "engineering"
> systems. The downside is that Axiom's type system does make learning
> the system harder. The upside is that Axiom is better both in terms
> of its mathematical structure and its programming structure. In the
> long term Axiom is where all of these system need to go.

Or, looking at it another way, Axiom is where users of other systems go
when they need the next level ;-).

If I may suggest this on the Axiom list, for now you might want to
start your students on Maxima or perhaps Yacas if you want a more
gentle introduction to CAS, and then move them to Axiom for more
advanced mathematical topics (maybe in upper level courses).  This has
two advantages (or I seem them as that, anyway). The first and biggest
is that your students don't get locked into one tool.  I've lost count
of the number of times I've seen students freeze up at something as
simple as a Mac interface, if they've never seen it before.  I feel
that any time computer systems are going to be used in an educational
setting, it is important to expose the student (eventually anyway) to
more than one tool and make sure they learn how to learn new programs,
because they'll likely be much better off in the future if they are
flexible about such thinks.  Second, introductory students aren't
scared off right away ;-). 

If you are an educator you doubtless have your own ideas or practical
limitations on such issues, and I'm sure others on the list will have
their ideas too (the zlc project might make this advice moot, except
for the part about exposing students to different tools).  Just thought
I'd add my two cents.

> In terms of student use Gilbert has defined a project to simplify
> the use of Axiom considerably (the so-called "zero learning curve"
> interface) which we are just beginning to work on. It is in the
> axiom--zlc--1 branch and is nowhere near ready to use. Myself and
> two graduate students in CAISS are working on it.

I've been curious about this Tim - I would have said Axiom's most
newbie-intimidating features (like being strongly typed) would be
pretty hard to hide.  Is there a discussion somewhere of how zlc
intends to achieve this?

> I can clarify what's going on but it isn't on the path to wisdom :-)
> 
> The windows branch does not yet have the graphics or the browser
> ported. Currently that code uses X11. We're looking at numerous 
> ways to get that code running in a more portable manner but it 
> hasn't happened yet.

I'm learning not to underestimate Tim, but based on other examples of
this kind of thing it may be a while.  Getting unix programs working on
Windows is always a little like riding a dune buggy - sheer excitement
with lots of hard bumps.  In the case of graphics, there are also a few
sheer walls which need climbing.

As a temporary measure, you might investigate the new Windows release
of gnuplot.  It's not integrated with Axiom, but it can generate decent
graphs from formulas.

> The February 2005 release of Axiom on Linux will have the full
> graphics and browser (which just came to life over christmas 
> cation) available.
> 
> Gilbert is looking to develop a linear algebra course that uses the
> browser technology with axiom as an experiment. That is also "in
> process" and not yet ready to show.

Cool.  An online course or a live one?

CY


                
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250