From: Kenneth Whistler (email@example.com)
Date: Fri Apr 21 2006 - 13:43:57 CST
Jukka Korpela asked:
> Why was (any) coverage needed _for MathML_?
> I can see the point in using
> mathematical italics letters and similar symbols _in plain text_, but
> isn't MathML supposed to be a mathematical _markup_ language?
> It might be more convenient, especially from the authoring point of view,
> to write the symbols simply as characters with code points of their own.
> But in a markup language, one _could_ also use markup for the same
> purpose, say, using <mi>h</mi> to denote mathematical italics "h".
Yes, but there were requests otherwise, both by the mathematical
community represented by the STIX project that worked with the
UTC for years on encoding mathematical symbols, and by implementers
of software that *uses* these characters, not only for mathematical
layout, but also for computational algebra.
There is no question that markup languages *can* mark up style for
mathematical symbols, just as they could for anything else -- but
this argument regarding the addition of the Mathematical
Alphanumeric Symbols was played out *years* ago, and there is
little point in rehashing it yet again.
> There's a potential future problem. Mathematicians keep inventing new
> symbols as they need them, using, say, Latin or Greek letters in some
> particular style (say, bold italic underlined and overlined - there are
> infinite possibilities). Will they all be encoded in Unicode?
The UTC and WG2 will burn those bridges when they get to them,
as they do for every new set of symbols brought in as candidates
The fact that the possibilities are "infinite" doesn't mean that
even mathematicians are so foolish as to invent infinite sets
of symbols to represent mathematical entities. After all, as for
every other use of technical orthography and nomenclature, there
is at least *some* intent here to be able to communicate with
others who can also learn and interpret your symbols. That's
why most mathematics does in fact continue to make use of
existing symbols and well known conventions for simply indexing
entities and symbols to indicate indefinite numerical generalizations.
So while people can cry Wolf! at the door here, in *practice* the
existing set of symbols will serve almost all purposes for
almost all mathematicians, and I expect that what we will see
will simply be a small trickle of additional symbols for
math in upcoming years and decades, as new symbols become
innovated and entrenched in actual practice. Small trickles
the committees are quite good at handling, actually -- so I
don't see a *real* problem here, as opposed to an *imaginary*
> Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
This archive was generated by hypermail 2.1.5 : Fri Apr 21 2006 - 13:47:29 CST