Proposed math coding scheme

[karl@cs.umb.edu (Karl Berry)]

    Alan writes:
    should we expect all document designs to be font-specific (in which case
    we can expect document designers to do the work of making sure the
    correct digits are used) or as font-independent as possible (so the same
    document should TeX as well as possible with different fonts)?
Fonts have to be specified in the (global) TeX document; but obviously
the same *textual part* of the document (i.e., not the macros/style
files) should be independent of the fonts used.  What this means, as far
as I can see, that the macro package designers have to 1) make it easy
to change fonts, and 2) be sure that the usual accent and other commands
-- \', \' and the like -- work for all fonts (or at least where at all
reasonable).

When the new math font encoding is published, I think it would be very
useful if TeX control sequence names were given to each character.
(Perhaps they already are; I still haven't been able to get a copy of
the original proposal.)  Likewise for the characters in the Cork
encoding.

    For example, [ is a Geometric in
    CMR, whereas in Monotype Baskerville it's a Humanist.
As you imply, Humanist-Geometric is a continuum, not a dichotomy, and
along several dimensions.  For the best typography, surely every
character is a Humanist.  For a short proof of some ad that no one is
going to see but me, every character might be a Geometric.

    b) The humanist symbols from cmsy, msam and msbm, including \dagger,
    \yen, etc. for use outside mathematics,

    c) If there's room (ho ho), this might be a good place to put the caps
    and small caps, universal currency, and the other bits and bobs that
    didn't make it into the Cork encoding.  Should Cyrillic or Hebrew
    letters used mathematically live here?
I don't understand why we want to put any non-mathematical characters
into the math fonts.  From a font designer's point of view, s/he is more than
likely only interested in creating a math font or a text font.  Very few
families will have both.  From a document designer's point of view, it
is simplest and cleanest if the ``math stuff'' is well-separated from
the ``text stuff'' -- since that's the way the existing fonts are going
to be arranged.

There aren't nearly enough positions in a 256-character encoding to
specify even all the useful math characters.  Why even consider adding
the bullet and dagger and so forth?
(I don't mean to sound harsh, I just don't get it yet.)

    Chris Rowley (and barbara?) suggested that we needed a math encoding
    for every font, not just italic.  
As you say, also, I think.  ``Math fonts are just like text fonts, only
different.''  That is, there's no reason to try to cram upright and
italic and bold and ... all into one font.  People may well want all those
variants -- at least whichever of them they can find.