From: Asmus Freytag (asmusf@ix.netcom.com)
Date: Thu Apr 14 2011 - 17:07:16 CDT
On 4/14/2011 9:02 AM, Murray Sargent wrote:
> Actually both TeX and Microsoft Office applications only have three script sizes: text, script, and scriptscript. So to handle math subscripts and superscripts, Unicode would only have to duplicate the current text-size encoding for the script and scriptscript sizes, which would only be adding about 200000 characters minus a few script sized characters that are already encoded. Needless-to-say, Doug's and Asmus's conclusions still apply :-)
Phew. :)
> Probably the closest one can come to encoding math in Unicode plain text is given in Unicode Technical Note #28.
>
Which, for those that haven't read it, is a light-weight semantic markup
scheme. It re-dedicates a few characters to serve as markers for scoping
and other aspects of semantic markup. Because the markup doesn't support
presentational features and because the semantics it carries are generic
("superscript", instead of distinguishing between "exponent" vs. "upper
indexation"), it can be small, and unobtrusive. (It's also not trying to
cover every unanticipated possibility, like TeX can).
The system in UTN#28 however, makes a good example how the correct
choice of atomic encoding on the plain text level can really empower
semantic markup.
A./
This archive was generated by hypermail 2.1.5 : Thu Apr 14 2011 - 17:12:08 CDT