On Sun, 2005-03-20 at 11:12 +0100, Kjell Rilbe wrote:

>
> David Johnson wrote:
>
> > I would say that this is a case for moving to UTF-8. That way, there is
> > never any ambiguity. Each piece of "national crap" has its own
> > datapoint that is universally understood and is distinct.
>
> You seem to be talking only about data. We're discussing (quoted)
> identifiers here.
>
> Kjell

Since identifiers are stored internally in a table (no different than
any other table), the distinction is a triviality. There is no
TECHNICAL reason not to allow this. There is, on the other hand, every
SEMANTIC reason to allow this.

Å is not an accented A with a circle over it. In Swedish and Norwegian,
it is its own character (different in each language) with its own
semantics, collation rules, and pronunciation. Likewise ß is not a
funny looking B. In German, it is called esstet, and is semantically
and syntactically identical to a "ss". In Greek, it is the letter Beta,
which is homologous to the English letter B.

What about the languages spoken by the majority of the world's
population ... The eastern tonal languages that don't represent properly
in any western alphabet at all? In Chinese, the difference between li
(kilometer), li (plum), and li (monkey) are impossible to represent in
ASCII unless one resorts to the Oxford Phonetic Alphabet, which does
allow for limited capture of tonality through the use of various
"accent" type markups. Vietnamese uses this mechanism.

Anglocentrism is a facet of our industry that must die, and the sooner
the better. UTF-8 is a sensible mechanism that is backward compatible
with ASCII, and allows representation of most living languages in their
native alphabet, without ambiguity.