[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Axiom-developer] MuPad under trouble

From: Nicolas M. Thiery
Subject: Re: [Axiom-developer] MuPad under trouble
Date: Wed, 9 Nov 2005 15:31:41 +0100
User-agent: Mutt/

        Dear Bertfried, dear Martin, dear Axiom developers,

> Nicolas: I'm copying this to you, since I'd like you to correct me
> if I'm mistaken...

Thanks Martin for keeping me updated. I completely second your
comments. Yes, the MuPAD-Combinat code is LGPL, so please go ahead and
reuse as much of it as you can! The whole point of this library is to
share as much as possible combinatorics code among as many researchers
as possible.

In fact, when we got to choose a platform for our Combinat package
five years ago, we have been pondering for a while whether we should
go for MuPAD or Axiom. We ended up going for MuPAD mainly because
Axiom's future was highly uncertain at that time (and also because
MuPAD had a larger user base and we already had some code under hand).

So, if our code could be ported to Axiom, that would be great!

Now to the technical part:

 - I have no clear idea of how much work that would be. We strongly
   use the Domain/Category/Axiom mechanism of MuPAD. It indeed is
   quite similar to that of Axiom, as it has been inspired by it!
   However, there are a few peculiarities, which might get in the
   way. That especially if the goal it not only to make a one-time
   port, but to keep the development synchronized on the two

   We also use quite intensively functional aspects of the language
   (lexical scoping, closures, ...), but I assume that's similar in

 - We will be very happy to provide technical support to anyone
   hopping in for doing the Axiom port; however, as long as MuPAD is
   reasonably alive we will concentrate on the MuPAD platform, and we
   won't be able to allocate serious development time ourselves to the
   Axiom port.

> To do the porting, it would be great if somebody knowledgeable of
> the bigger structures in MuPAD (i.e., what does a domain, a category
> look like) would write a bunch of regexps that would save us a bit
> of typing...

Ok. The efficient way to do that would be to have some MuPAD-Combinat
guy meet physically with some Axiom guy and have a one/two day coding
party on the subject. If we want to do more than a one-time port,
maybe the right approach would be to have a MuPAD parser in Axiom
(with possible legal issues to check).

> > The combinat people integrated a tremendous amount of code from
> > symmetrica, partly SCHUR and forged from these special purpose
> > (few used?)  programs a good tool.

Thanks :-) Do you mind if I post your comment on our "user
testimonials" wiki page?

Best regards,
                        Nicolas, for the MuPAD-Combinat group
Nicolas M. ThiƩry "Isil" <address@hidden>

reply via email to

[Prev in Thread] Current Thread [Next in Thread]