W3C home > Mailing lists > Public > www-math@w3.org > July 2002

Re: Statistics and MathML

From: William F Hammond <hammond@csc.albany.edu>
Date: 24 Jul 2002 09:43:01 -0400
To: W3C MathML Discussion <www-math@w3.org>
Message-ID: <i7ptxdnswq.fsf@hilbert.math.albany.edu>

Michael John Haag <mjhaag@owlnet.rice.edu> writes:

> I was just wondering if anyone has considered the use of statistics in
> MathML.  In particular we are looking at marking-up functions such as
> autocorrelation, probablity density/distribution functions, etc...  Has
> anyone considered implementing these functions and other STAT notation
> correctly in MathML via Csymbols or some sort of "statML?"

If you're looking for ways to extend what is presently available for
present use, an answer will depend on the method used to create your
content.

With GELLMU one can sketch future document type names in terms of
present names using \newcommand and then once the new names become
available, one has the option of backing out the newcommands while
leaving their calls.  However, at this stage I know no comfortable
document type for authors.  Writing xhtml+mathml itself is only less
painful with gellmu than writing it raw.

That could go away quickly if somebody had the energy to write a
translator from gellmu article to xhtml+mathml with good guidelines
about when an author needs to provide invisible logical groups and
what attributes should be used where.  (Meanwhile the article to
classic-html formatter does, I think, a reasonable job of handling
math accessibly.  For example, w3m seems to be OK with it.  See
mathbench, confrac, and f356g in the gellmu examples archive.)

It will be at least a year before I am likely to have something in
this direction because of (1) other demands on my time and (2) my
uncertainty about how LaTeX authors will actually be willing to
undertake relatively small departures from current practice.

For my indecisiveness about (2) anything done in the same direction by
others might give me clues about what should and should not be done.
For example, references to disciplined LaTeX source that goes well
through the MathML track of Eitan Gurari's tex4ht might be helpful.

                                    -- Bill
Received on Wednesday, 24 July 2002 09:43:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:51 GMT