[XeTeX] The future of XeTeX

[mskala at ansuz.sooke.bc.ca]

On Wed, 1 Aug 2012, Ulrike Fischer wrote:
> Perhaps (the discussion is rather long). But you obviously don't
> accept my conclusion that one possible solution is to reduce the
> complexity of the script. You are only looking for the people who
> should write all this complex code.

Language is inherently political, and telling people to change their
language to suit the computer is really asking for trouble.

However, something that might fly better and addresses similar issues
would be to say:  requiring the typesetting system to build in per-script
support is a losing game because it requires the builders of the
typesetting system (who will be experts on computing, not on ALL the
scripts of the world) to learn ALL the scripts of the world.  It's also a
political problem because some scripts, or some forms of some scripts,
inevitably won't make the list of "all" scripts and will be
disenfranchised as a result.  So:  this per-script knowledge should be
moved from the typesetting system to the font, and then it becomes the
responsibility of the font designers who more reasonably can be expected
to be experts on their own scripts, and then nobody needs to be an expert
on ALL scripts, and unforeseen scripts can be easily added just by
creating new fonts.

That is the line of thinking that would favour Graphite (a general system
for defining complex scripts inside fonts) over OpenType (which requires
each script to be defined in the typesetting system, outside of the font),
and it should be acceptable both to people who want the technology to be
easy to build and to people who want the output to look right.

-- 
Matthew Skala
mskala at ansuz.sooke.bc.ca                 People before principles.
http://ansuz.sooke.bc.ca/