- From: David Carlisle <davidc@nag.co.uk>
- Date: Fri, 30 Jan 2004 12:21:20 GMT
- To: Stephan.Semirat@ac-grenoble.fr
- Cc: Vincent.Quint@inria.fr, www-math@w3.org
> you only stretch parenthesis with what they contain, do you ? That is not the model used by MathML, the model is that you stretch to the surrounding element not the containing one. TeX of course only has the "surrounding" model, which means that it has to invent a special syntax for null delimiters to allow expressions with just one bracket stretching, {a x={b {c and it has real problems with syntax such as ( x | y) with all three fences stretching. the MathML model seems slightly strange at first but it deals much more naturally with both those cases. > So my view is that Mozilla has made the good choice... But in this case it wasn't mozilla's choice to make, the choice was made by the designers of MathML1 (not me:-) David -- http://www.dcarlisle.demon.co.uk/matthew ________________________________________________________________________ This e-mail has been scanned for all viruses by Star Internet. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ________________________________________________________________________
Received on Friday, 30 January 2004 07:59:42 UTC