|Again, I suspect most readers of this list aren't all that interested in
|this topic
On the contrary, this is one of the most fascinating discussions I've
seen!
Assuming Glenn is right (he certainly sounds very authoritative), and
we can indeed map every standard charset to ISO 10646 (possibly with
the help of HTML's language attributes in the case of CJK), and given
the fact that there are clear advantages to having a single unified
character set, can't we then simply close the issue here and say that
HTML can express every character in ISO 10646 plus every character for
which we define an SDATA entity, and no more?
If people then want to use other character sets, they simply have to
use something other than HTML. I know there are many more characters
then there are in Unicode, but wouldn't it suffice to define character
entities if/when they're needed? For some alphabets (Glagolithic, for
example), such entities already exist.
Bert
--
Bert Bos Alfa-informatica
<bert@let.rug.nl> Rijksuniversiteit Groningen
<http://www.let.rug.nl/~bert/> Postbus 716, NL-9700 AS GRONINGEN