Re: MathML to HTML+gif?


You wrote:

> I think you misunderstand my concern. It is not primarily buggy renderers,
> that I am worried about.  I am more worried about buggy emitters of MathML.
> It seems that it will take a while before there MathML support will be
> built into mass market browsers. It appears likely that software to
> emit MathML will probably appear first on the market, as a selling point
> for various kinds of eg. word processing software.
> What if say, eg. MS Word produces buggy MathML? Then MS would probably
> support these bugs in its browser, etc.

You certainly make a valid point.  I still suggest that in the
universe of buggy markup emitters, people are rather more culpable
than machines.  Of course, they do have the advantage of committing
their sins in a relatively random fashion, so no one would be tempted
to build a renderer to accomodate any particular error.

Nonetheless, to me, it seems like the most practical way to test a
MathML emitter is to have a collection of good MathML
renderers/validators to check what it emits.  That is why I was
focusing on test suites for renderers.

Finally, I certainly agree it would be a blow if a major player like
Microsoft entrenched a faulty implementation of MathML in IE and MS
Word.  However, in that particular instance, I don't think that is at
all likely to happen, since Design Science (the company that makes the
MS Word equation editor and MathML exporter) and Microsoft itself both
have representatives on the W3C Math WG, and are very much committed
to doing it right.  In particular, Murray Sargeant, the Microsoft
representative, has a long history of work on Unicode, and is a very
able and impressive defender of the advantages of standards.

For once, the relative unimportance of MathML from an economic point
of view is a help -- no one has much of a motive to try to co-opt
MathML for proprietary advantage.  Quite the contrary.  Almost
everyone I have talked to in the math software biz sees the ability
to effectively intercommunicate with other software as a way of
expanding their markets.  


Robert Miner                          http://www.webeq.com
Geometry Technologies, Inc.           email: rminer@geomtech.com
                                      phone: 651-223-2884