Re: Octave and Guile?

[Mikael Djurfeldt]

"John W. Eaton" <address@hidden> writes:

> I'd really like to know whether people think that Guile will have any
> real impact any time soon.

The interest for Guile is currently increasing.

As you probably already know:

* Guile-1.4 is faster and nicer to work with than Guile-1.3, and we
  have already made significant improvements after the 1.4 release.

* The manual is now well under way, and there will probably be a
  pre-release of it soon.

* We now have a specification for translator support in Guile.  It
  only needs small refinement before we can go ahead implementing it.

* Keisuke Nishida has implemented a VM for Guile and Dan Crettol is
  working on the next generation of his QScheme VM and has joined
  Guile.  We plan to replace the current evaluator with either of
  these engines.  The fact that Guile then probably will be the
  fastest Scheme interpreter will increase popularity a lot.

* Michael Livshin has just replaced the GC with a card-based one.
  It's possible to implement generational GC in this infrastructure.

* GOOPS will soon be integrated into libguile.

* We are pushing for having some kind of Guile code library on the net
  where Guile users can share code.  This is also one of the points
  which will boost Guile popularity.

> Could Octave provide some momentum for Guile, or would it just be
> seen as a minor development?

I think Octave would provide a lot of momentum for Guile.  I see it as
very important.

(Personally, as a user, I also think Guile would boost the usefulness
 of Octave.)