Re: [Texmacs-dev] Efficiency of scheme implementation

[Joris van der Hoeven]

On Wed, 18 Feb 2004, David Allouche wrote:
> On Tue, Feb 17, 2004 at 11:01:03PM -0500, Michael Graffam wrote:
> > > Thanks for the pointer. This will probably still take some time though.
> > > In any case, I feel that the Scheme interface should be modified in such
> > > a way that it is reasonably easy to plug-in abitrary Scheme 
> > > implementations.
> >
> > Being able to easily plug-in arbitrary Scheme systems would be great.
> > It would be relatively easy to create a Scheme dialect in Common Lisp
> > (the pseudoscheme implementation might be a good start) that would be
> > sufficient for TeXmacs. A Lisp backend would provide the needed foundation
> > for a measure of Emacs compatibility, as well as making TeXmacs
> > that-much-more flexible.
>
> One thing to keep in mind is the more portable you want it, the least
> you can rely on non-portable library functions... note that pure R5RS is
> a really poor dialect for any non-trivial amount of programming.

I think that this is actually a good thing.

> For example, texmacs relies on the the Guile module system,
> define-macro, procedure-properties, and maybe on a few other
> guile-specific features which might or might not have equivalent on
> some other given scheme system.

Yes, and I am actually not quite happy with the current module system.
In fact, it does not really well integrate with the DRD system,
the menus, etc. After some more urgent reorganizations,
the module system has to be rethought out quite carefully.