- From: Neil Soiffer <soiffer@wri.com>
- Date: Wed, 12 Aug 1998 10:20:13 -0700
- To: davidc@nag.co.uk
- Cc: www-math@w3.org
>Isn't raising fractions just a style issue to be decided by the render >or style sheet, that is, couldn't the mathml > ><mrow> > <mi>a</mi> > <mo>/</mo> > <mi>b</mi> ></mrow> > >be rendered as: > > (like 'a/b', but letters are smaller then regular and 'a' is located > higher then 'b') > >without any special markup, if that is the style you want for inline fractions? > >(although that would have all the horrible features of tex's \over >command: requiring an arbitrary amount of backtracking once you find the >/ in order to raise the a. However that backtracing is needed anyway >because / is specified as stretchy in the operator dictionary, so its >size depends on the size of a. (Which incidentally means, I believe, >that DSSSL (never mind XSL) is not powerful enough to render MathML, >which is a shame.) > >David No backtracking (in the conventional sense of the word) is needed for rendering stretchy operators. It is, as specified in the recommendation, a two pass linear operation: find the size of all non stretchy children, tell all of the stretchy children to stretch to that size. At a MathML meeting in INRIA this spring, we were told (basically) 'XSL will never be powerful enough to render MathML'. To be precise, I think what was said was more like: 'you are welcome to propose optional flow objects for XSL to support MathML, but it is very unlikely that they will be implemented by the major browser manufactors'. Neil Soiffer
Received on Wednesday, 12 August 1998 13:20:57 UTC