axiom-developer
[Top][All Lists]
Advanced

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

Re: [Axiom-developer] Re: hyperdoc


From: Bob McElrath
Subject: Re: [Axiom-developer] Re: hyperdoc
Date: Fri, 14 Jan 2005 17:50:55 -0800
User-agent: Mutt/1.5.6+20040523i

C Y address@hidden wrote:
> Just so I'm clear, are we discussing adding the ability to output
> MathML and perhaps read it in, or are we talking about making OpenMath
> the native language/mode/whatever of Axiom?  MathML in/out I can see as
> being about the same as TeX in/out - desirable in some situations, but
> not for internal CAS use.

For the immediate future, I'm talking about putting MathML on the web
(however that is done).

However there are significant implications to choosing to handle XML
within axiom (as I outlined in previous email), that affect the 30-year
goal.

> a)  MathML is primarily useful as a way to export CAS worksheets (or
> whatever you're doing your math in) for web display.  It shouldn't be
> an integral part of the calculation process.

Agreed.  But will that always be the case?  Are we *sure* OpenMath will
flop, and the ideas it represents?  (and stay flopped for 30 years)

Let me make the possible decision chain explicit:

    1. export CAS to web => make axiom write MathML directly

    2. add lisp XML library to axiom to make (1) easier

    3. OpenMath is now easier, because we have XML already

    4. A browser-based worksheet program/editor appears (a la mozilla)
    that can generate OpenMath or content MathML to feed to a CAS.
    (e.g. it has palettes like Mathematica/Maple)

    5. we add OpenMath/content MathML as an input format

I'm just saying that design decisions that tie us to XML are probably a
good idea, and probably have a better future than design decisions that
eschew XML in favor of TeX.

> c)  In general, display of 2D equations in confined space is a hard
> problem.  TeXmacs does some line breaking, and so does wxMaxima, but
> I'm not sure how powerful the algorithms are.  Any 2D equation based
> GUI solution needs to be considered carefully though.

I would think this could be done in the browser?  OpenMath or Content
MathML plus an XSL stylesheet could do it, presumably...

BTW here is a critique of OpenMath by Fateman:

    http://www.cs.berkeley.edu/~fateman/papers/openmathcrit.pdf

and a discussion of his criticism starting here:

    http://lars.math.fsu.edu/archive/om-announce/msg00096.html

He doesn't like it so much...and brings up a number of good points.

Cheers,
Bob McElrath [Univ. of California at Davis, Department of Physics]

    "Faith: Belief without evidence in what is told by one who speaks 
    without knowledge, of things without parallel." -- Ambrose Bierce

Attachment: signature.asc
Description: Digital signature


reply via email to

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