[Axiom-developer] Re: GCL on windows.

[Bill Page]

Tim,

On Wed, 27 Nov 2002 16:29:38 -0500 you wrote:

> ...
> > [Bill] 
> > The biggest difficulty that I see in MinGW is the lack
> > of the X-windows environment which is (more or less)
> > unique to Unix (and part of Cygwin). Microsoft Windows
> > itself *is* a windows environment but very different
> > from X-windows. So porting Unix graphical applications to
> > Microsoft Windows is currently rather difficult. But their
> > are some people working on methods to make this easier in
> > the future.
> > 
> > After several tests and trials, I think MinGW/Msys (rather
> > than Cygwin) is the best choice for building the Windows
> > version of Axiom.
> 
> Ok. I'll defer to whatever choice you find appropriate.
> The graphics portion of Axiom currently amounts to the
> hypertex browser and the drawing software. The interpreter
> doesn't really depend on anything since it was developed
> to run on a text-only mainframe. Getting interpsys and
> the algebra running is the first goal.
>

Yes, of course.
 
> The windows version of Axiom used to use IBM's
> TexExplorer software but that was not released. I'll
> cruise around the web for some combination of tex,
> graphics & windows software. We can't be the first
> people to need an open-source version of this.
>

Well we have been discussing TeXmacs as a graphical
interface, right? TeXmacs does not currently run in
native windows but a version is planned. A user interested
in a graphical tex front-end to Axiom running on a PC could
still install Cygwin and run TeXmacs to interface with
Axiom running in windows native mode.

Currently, as I understand it, TeXmacs has not "off-the-
shelf" facility to display graphical output from tools
such as GNUplot, etc. But I have seen this discussed
on the TeXmacs developer list.

> Magnus has a tcl/tk front end. I believe that tcl/tk
> software will run on windows. Do you know if this is
> true?

Yes. But these tools are a little "dated". If changes
are needed to Axiom then we might consider other
alternatives. I can check around and let you know what
I find.

If our goal is just to support the existing Axiom graphics
capabilities in a platform independent way, then perhaps
simply re-coding that part of Axiom to interface with an
open source graphics package is all that would be needed.

> Perhaps we can steal the front-end as it is a stand-alone
> piece of software.
> 

What are your current thoughts about TeXmacs? Personally
I still think it is a good match for Axiom and I would be
prepared to devote some effort to helping to extended
TeXmacs capabilities where these would better meet the
requirements of Axiom.

I think it would be good to avoid having to develop an
Axiom specific front-end (even if it is "borrowed" from
some other open source package), if at all possible.

Regards,
Bill Page.