Re: Implementing Complex Unicode Scripts

Date: Tue Feb 20 2001 - 13:22:52 EST

On 02/20/2001 11:19:48 AM Antoine Leca wrote:

>Yes. Font technologies does not allow things like Nagari or Sinhala
>to operate by themselves, they need some assistance from the underlying
>This is the current state of art, one may hope it will change in the

I don't know that we should hope for a change from having an underlying
platform provide support for rendering of scripts like Devanagari or
Sinhala. What we should hope for are (i) that such support become available
on all underlying platforms, and (ii) that applications begin to make use
of that support.

We might also hope for a change in how such support is actually implmented.
One approach requires binary code that implements support for specific
scripts, with the result that extending support for an additional script or
adapting support for variations in particular writing systems must be done
by (typically) C/C++ programmers and involves a re-release of the entire
engine. This is the current situation, for example, with Uniscribe. Other
approaches might include

- that the single monolithic engine is replace by a more modular system in
which single rendering management component manages pluggable
script-/writing system-specific engines (which means that capability can be
extended without requiring a complete recompile and re-release), and

- that rather than having script-specific engines you have a generic engine
that can process descriptions (such as are possible with GDL) that can be
prepared without requiring something like a C/C++ programmer.

- Peter

Peter Constable

Non-Roman Script Initiative, SIL International
7500 W. Camp Wisdom Rd., Dallas, TX 75236, USA
Tel: +1 972 708 7485
E-mail: <>

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:19 EDT