From: Antoine Leca (Antoine10646@leca-marti.org)
Date: Wed Aug 24 2005 - 07:17:01 CDT
On Wednesday, August 24, 2005 1:40 AM Richard Wordingham wrote:
> The mechanism is clunky, but it exists.
Yes. But if think about it to handle say Tamil (one of the most simple Indic
scripts when it comes to rendering), you'll end up with buggy slow-drawing
fonts.
When you step up to Nagari, the projected size of the font become a real
problem, the probability for bugs is too near from 1.000 to be allowed to
provide another estimation, and it is simply too much work involved to be
able to have a reliable estimation for performances.
Back in 1995-6 (and please remember the computers' performances then), MS
did have those two scripts to handle (from TUS 1.0, volume 2), and did have
the TTO technology (the basic GSUB etc.)
The result of the thinkings is the current OpenType for Complex scripts
specifications.
Antoine
This archive was generated by hypermail 2.1.5 : Wed Aug 24 2005 - 07:17:55 CDT