Precomposed chars, PUA, and the future

From: David J. Perry (
Date: Sun Oct 10 1999 - 18:26:29 EDT

I read with great interest the recent threads on precomposed characters and
such. I now have a much better understanding of reasons for avoiding
precomposed characters, and why it is unlikely that any more will be added
to Unicode. Thanks to all who contributed!

This leaves unanswered, however, the question of what those of us who are
looking to Unicode to meet the needs of our languages/scholarly disciplines
should do right now. In my case, I'm working on a proposal to use the
Private Use Area to complete the set of characters required for classical
Latin and classical Greek. A number of these could be handled with
combinations of character + combining diacritic (e.g., y-breve); but, due
to the limitations of current word processor/font technology, I had planned
to set them up as precomposed forms in the PUA. I am now rethinking that idea.

So my question boils down to: when will we have software that that is
accessible to a large number of users and that can handle combinations such
as SMALL LETTER Y followed by COMBINING BREVE ABOVE? Win2000 is supposed
to support OpenType, but I have heard that it will be up to the individual
application to implement (or not) the various features. If I create an
OpenType font with appropriate positioning tables to make sure that all my
combining accents end up in the right spot, and widely-used apps don't
support this feature (except, perhaps, if the language being used is one
that _requires_ it, like Devanagari), then I'm better off sticking with my
original precomposed scheme.

I'd appreciate any help with this conundrum. -- David

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:53 EDT