[XeTeX] Word 2007 Math

Hans Hagen pragma at wxs.nl
Tue May 30 09:31:18 CEST 2006

Venkatesan SK wrote:
>     Adam Twardoch wrote:
>     > * Microsofted consulted with Donald Knuth on the development of their
>     > new math typesetting engine 
> Even Bill Gates was a different guy when he was young and a programmer.
>     >
>     and also, if i'm right, latex's author (lamport) has been working for
>     microsoft research for years now -) 
> It is good to have some fare competition but with monopolies you are 
> talking about unfair competetion....
says the texie, using a program that has dominated the math typesetting 
world for 25 years -)

imagine what it must have been for authors of amstex, lamstex, inrstex 
and maybe more interesting variants of the early days to compete with 
latex originating at one of the (in those days) huge hardware/software 
companies [dec];
even if context supports most of what latex math does (which is rather 
plain anyway), maybe even more configurable, latex math will always be 
considered 'a standard' and has a dominating monopoly as well; the 
original idea of DEK, that anyone would cook up his / her own brand of 
tex, was quickly killed by organizations and publishers who demanded 
this one tex dialect; it's all so relative ...
> TeX uses LISPish curly bracket notation, XML uses angle bracket 
> notation, other than that I can't see anything new. Cosmetics are very 
> important these days...they don't care much about content... it is so 
> sad we have to hear lecture on quality typesetting from M$...
actually, tex does not really uses a competely structured input; in the 
'article' they take \frac as an example, and one of the reasons for 
\frac being there (as macro) is that it replaces the\over primitive 
which is one of the things that is meant for sequential input but at the 
same time forces multiple passes, look back/ahead, complicates font 
stuff, and such; without that one, live would have been easier; this is 
also why abc/123 as sequential input looks nice but can become a parsing 
(and quality typesetting nightmare);

even more dangerous is the fact that users can choose from the whole 
range of unicodes and believe me, we've seen, in one document many 
different *dashes being used, greek and non greek being mixed, chars 
that looked like the one needed (in this or that font) being misused for 
...; or presentation forms ending up in the document source (like fi); 
or take the superiors: one will never really know if a superscript 12 is 
meant to be used or sup1 sup2 (because the author wanted it to look that 

on the other, it's not that much different from what authors using tex 
do ... when ons submits an article to a journal in tex code, one should 
not be surprised when in the copy-editing stage the whole lot is rekeyed 
in a consistent format without the author noticing; so, in the end, it 
does not matter much if one submits an article in ms-office format or in 
tex-format -)

an other aspect we should not neglect is that, just as those who had to 
write math were delighted that tex came around (replacing the typewriter 
mixed with handwriting), there will be many happy authors doing simple 
math who will be delighted that they can enter it directly in word 
without the need for plugins, endless click and point, reading manuals, etc
concerning the nearly plain text math encoding ... (where the url 
pointed to) ... it's

(1) not a standard but an initiative of a ms employee (and never will be 
a standard i think)
(2) it's mostly dealing with simple [look and feel] math (as is pres 
(3) my guess that the amount of research involved (or at least the 
amount of complex math being typeset developing the scheme) is not that 

but ... it would not be the first time that texies would end up being 
forced to follow such 'standards' simply because there is not much 
progress in their world (how long did it take to get math into unicode, 
how badly did/does mathml take cultural aspects into account, how many 
open type math fonts originate in the tex world (ok, there is now work 
in progress with regards to open type fonts), how often were new 
initiatives related to math more or less killed (i still remember 
reactions to an Nath (package) presentation at a tug conference: not 
interested, even if it does a better job, because we're doing it 
this/that way for 20 years now and it's not going to change')

so, instead of focusing on what we don't like about ms office math, we 
should look at where the strong points of (future) tex's lay and how we 
can let those worlds work together


More information about the XeTeX mailing list