unicode support (and misc encodings)

[Denys Duchier]

Hi Maarten & Vladimir,

as Peter suggested, your very valuable efforts to add unicode support to Mozart
need to be presented through one or more MEPs.  I have looked at the patches
that you have made available (again thank you SO MUCH for that work), and I have
to give you some unsolicited (and quite possibly unwelcome) advice:

You need to break down your changes into smaller units: there is NO WAY that we
will accept such (in|per)vasive changes into the official sources in one block
and without very detailed (and convincing) explanations.  We need to understand
your goals, your plan, the incremental steps to get there, and the design
principles that back them up.

This is why I originally suggested a much more modest enterprise: namely,
writing a library for encoding and decoding (for various coding systems).  I
wrote one for unicode/UTF-8 (based on the "on-demand batch" design described in
an another message) and I imagine others have too.

If you want to push for unicode support, it is up to you to initiate and sustain
a discussion on the subject; and an important prerequisite to getting people
involved in such a discussion is to break it down into modest manageable
chunks: one thread for one (small) subject.

To get things started, I'll follow up with a thread on unicode/UTF-8: we should
probably discuss how it is supposed to be connected to the Open module.

Cheers,

-- 
Dr. Denys Duchier - IRI & LIFL - CNRS, Lille, France
+33 (0)6 25 78 25 74    http://www.lifl.fr/~duchier/