Masataka Ohta reminds us that:
The problem is that, from the view point outside of Europe and US, ISO
10646 is merely a poor extension to ISO 8859-1.
It assignes a single code point to different but similar characters
in Japan and China.
So, please don't say "international" when what you mean is merely
"intereuropean".
I think it's terribly important that we keep this in mind as we develop
standards for the future, and design, insofar as it's possible to doso, for
the long-range goal of a truly international -- not just intereuropean --
network, though it will probably be necessary to miss that goal for
pragmatic, interim solutions for a while, just to keep moving on some of
the issues for which there are no clear solutions.
I must confess that I like Rick Troth's idea in this context, though I
certainly don't understand its practical implications yet:
... I say we scrap the 16-bit
stop-gap solution and go directly to 32-bit and then start looking
toward bit-unconstrained (bit-free?) representations.
Jim