The ISO TR 9573-11 DTD fragment is in use in the U.S. Department of Energy; we
have conversion routines (in OmniMark) to TeX at our site (since math in HTML
isn't finished, we do math on the Web by SGML-->TeX-->PostScript-->GIF).
As Paul says, most routines for handling math are compromises. After all, the
purpose of mathematics is in many cases to develop new relathioships and
semantics, and you can hardly codify that sort of thing in a DTD or any other
fixed document that must be prepared and shared in advance. So mathematical
systems tend to be rendering-based systems, a tradition that goes back to the
compromises in TeX and, before that, in eqn (both of which are typographic
systems, not information management systems, anyway).
Anders Berglund, whom Mike Cowlishaw remembers at ISO CS, is now at Electronic
Book Technologies (where his wife, Sharon Adler, editor of DSSSL also works:
alb@ebt.com, sca@ebt.com).
Dr. James D. Mason
(ISO/IEC JTC1/SC18/WG8 Convenor)
Oak Ridge National Laboratory
Information Management Services
Bldg. 2506, M.S. 6302, P.O. Box 2008
Oak Ridge, TN 37831-6302 U.S.A.
Telephone: +1 615 574-6973
Facsimile: + 1 615 574-6983
Network: masonjd @ ornl.gov