[EMAIL PROTECTED] >this may be difficult to achieve in practice, because TeX >conversors reading TeX sources are unable to provide correct MathML markup >for prescripts.
Conversion to MathML is obviously more difficult because the base has to be found and encoded explicitly. Still, I do _not_ say that conversion from TeX to HTML5 will be trivial in all cases. >[classes:] "vector" or "Hilbert-vector" [or] "ket". >I would leave [the choice] to authors until that a generic >semantic markup was achieved, proved to be consistent and powerful and >then used by authors. With all the different concepts and notational conventions that exist in different scientific fields and mathematical disciplines, no such thing as a generic semantic mark-up is likely to appear any time soon. I tend to believe that you agree. >Font styles would be specified via CSS styles. ><span class="bold"> is not defined in HTML >text, the font bold property is defined in CSS and you can call it via a >CSS rule applied to a class you can define. For example Spanish authors >could prefer <span class="negrita"> The combination of author-defined classes and CSS styling obviously makes sense in many ways. Still, I am a bit reluctant when it comes to encoding the right font using the class attribute. Perhaps a profile would make it acceptable, as suggested by Michel Fortin. >Also "tag" is not usual in French texts. I am afraid I do not understand what you mean. -- Andersen
