[XeTeX] Free download: Microsoft Office 2007 beta 2

Hans Hagen pragma at wxs.nl
Mon May 29 22:57:16 CEST 2006

Hi Adam,
> * the new math typesetting engine uses Unicode, which is an 
> international standard (ISO 10646)
> * the new math typesetting engine uses MathML, which is an international 
> standard from W3.org and XML-based
> * the new math typesetting uses OpenType, which will soon be an 
> international standard (as ISO Open Font Format)
a few remarks: 

(1) mathml, esp presentation mathml is not that usable for complex high quality typeset math (for one always needs directives to control over look and feel; content mathml and open math provide more consistency (rather important) and control; we can proces that kind of stuff now for over five years using tex (nativ eparsing) but so far only a few people showed interest in that (we use it in a few projects esp educational math); open math is the most flexible but also demands a configurable engine (we use open math in one large scale [pet] project) 

(2) there are cultural aspect involved, i.e. different renderings of math constructs; the same for different usage of math (secondary school, high school, university, as well as locales) 

(3) the sequential unicode input scheme is not that different from e.g. (what we call) calculator mathm which can be handled by tex using parser macros or otherwise; btw, i demonstrated this at bachotek; again, this is sufficient for <= high school math, but beyond that ... ; anyhow, thanks for the pointer, i'll see if i can get this UNPTEM working 

(4) once matrices and bracing comes into play, all inputs become more verbose, esp when numbering and alignment gets involved; no matter what input one chooses, the problem never changes -) 

now, the interesting question is: how will mathematicians react on this; i still remember some of the reactions when i presented the context mathml parser (texies are very hooked to tex math input) 
 5 tokens: $x+y$ 
10 tokens: <m>x+y</m> 
 7 tokens: [[x+y]] 
 3 tokens: x+y 

at first sight the x+y looks the most simple (more complex formulas get closer numbers), and can still be recognized as math, but how about 

 3 tokens: $x$ 
 8 tokens: <m>x</m> 
 5 tokens: [[x]] 
 1 tokens: x 

however, how do we know that the x is math (font switch = a few clicks; math selection (also a few clicks), so ... this is why mathematician like the $x$ 

In the end it would be nice to see multiple inputs being accepted (which is btw what we do in most of our projects); 

In any case, by being a rather self centered group of users, texies has lost much of the lead in such matters. 

Again, thanks for the pointer, 


                                          Hans Hagen | PRAGMA ADE
              Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
     tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com
                                             | www.pragma-pod.nl

More information about the XeTeX mailing list