Re: new Emacs HELLO file??

[Kenichi Handa <handa@m17n.org>]

In article <873a762ylx.fsf@mail.jurta.org>, Juri Linkov <juri@jurta.org> writes:

> > As iso-2022-7bit has designation sequences for each
> > character set, Emacs can use that information to add proper
> > `charset' property, and that property infulence the font
> > selection.

> The purpose of the HELLO file is to demonstrate Emacs multilingual
> capabilities.  It currently does this using iso-2022-7bit.  Does this
> mean that UTF-8 is not the best coding system for multilingual texts
> in Emacs, because UTF-8 has some limitations (doesn't provide the
> proper font selection, loses cjk variants), so iso-2022-7bit is the
> preferable coding system for multilingual texts in Emacs?

At least for CJK characters included in legacy character
sets (e.g. JISX0208, GB2312, etc), iso-2022-7bit is still
better than utf-8 if we concern font selection.

> Is it possible
> to improve the UTF-8 support in Emacs since most files use it nowadays?

For that, we must use language tags, but they are strongly
discouraged by Unicode.  Unicode doesn't concern font
selection problem.  It says that it is a task of higher
level information, for instance, that is provided by XML's
language tag.

---
Kenichi Handa
handa@m17n.org