Janko Stamenovic <email@example.com> wrote:
> I just demonstrate to you that the REAL standard must take care of
> everything, not only parts of the issue. If we have to know about language
> tags, we either should have incorporated them in the code or made the
> code/standard good enough to take care of it.
Unicode by itself is insufficient for this kind of thing - even if language
tag characters are included, to display the characters in a culturally
acceptable manner you need fonts tailored to specific languages or something
like AAT or OpenType fonts with language specific variant glyphs and
software that can use them. If you understand how AAT and OpenType fonts
work many of the "problems" with Unicode disappear. There probably should be
some explanatory material in the published Unicode standard and on the
Unicode web site explaining, in a generic way, how character to glyph
mapping and substitution works.
People concerned only with "simple" Latin or Cyrillic script may believe
that they can get away without the kind of complex rendering systems like
those required for e.g. Indic scripts. However as more and more examples are
provided about language / culturally (and other) specific variations for
these "simple" scripts, it seems to me that they are actually far more
complex scripts than is generally assumed.
If people want to build Latin or Cyrillic fonts which are acceptable in all
the different locales and for all the languages where these scripts are used
they are going to have to contain some language / country specific glyph
mapping and the necessary alternate glyph forms.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:58 EDT