axiom-developer
[Top][All Lists]
Advanced

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

RE: [Axiom-developer] steps towards better TeXmacs interface


From: Bill Page
Subject: RE: [Axiom-developer] steps towards better TeXmacs interface
Date: Fri, 26 Aug 2005 16:28:40 -0400

On August 26, 2005 4:00 PM Martin Rubey wrote:
> 
> C Y writes:
> > I think I know what Andrey is thinking about here.  Let's 
> > say we have the following expression:
> > 
> >                               2
> >                              b  + 2 a
> >                              -------- + c
> >                                 d
> >                              ------------
> >                                 f + e
> > 
> > and we want to select:
> > 
> >                                 2
> >                                b  + 2 a
> >                                -------- 
> >                                   d
> > 
> > as our input in the next line. 
>

I think that is a very good and entirely plausible example.
 
> I'd say that this is not the way things work in Axiom. There 
> is a hack on MathAction that does this kind of thing.
>

Martin, could you be more specific? What "hack on MathAction"
are you referring to?

Although I think I see your point about "not the way things
work in Axiom", I also think you are being too strict when
it comes to user interface. Axiom is designed as a deeply
nested series of levels. A the user interface level a lot of
"less strict" things are done in the name of being more
"user friendly". This is a difficult problem and I think
achieving a truly effective user interface is still more a
matter of art (and luck) than it is a matter of good design
or engineering.

> But please keep the following in mind:
> 
> Axiom is designed in a way that you can, for example, replace 
> the domain EXPR ? with something entirely different, keeping
> only the interface the same. Of course, coercing to Sexpression
> is part of the interface, but I would suggest that you don't
> work against Axioms philosophy.
> 
> Note also that "hidden" type information is crucial:
> 
> (13) -> 3*(y*x::XPOLY INT)
> 
>    (13)  y x 3
>                                                     Type: 
> XPolynomial Integer
> (14) -> 3*y*x
> 
>    (14)  3x y
>                                                      Type: 
> Polynomial Integer
> 
> (XPOLY: variables do not commute)
> 

We should not forget that Numerical Algorithms Group put a lot
of work into Axiom to produce OpenMath output. One of the goals
of OpenMath is to encode precisely this kind of information with
an aim to being able to communicate mathematics in full generality
between different computer algebra systems. And of course this
could also include between a compute algebra system and an
intelligent editor/user interface. All of the NAG OpenMath code
is availabe but has not (yet) been integrated into the open source
version of Axiom.

Regards,
Bill Page.






reply via email to

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