I am the convener of the ISO working group (ISO/IEC JTC1/SC2/WG2) that
developed the 10646 standard and continues to work on it right now.
Really? Then, I have a question. What, do you think, "Universal" means?
First, we believe that the characterization of ISO/IEC 10646-1 as needing
external profiling for practical use is incorrect.
Unfortunately, you are wrong.
There are several
companies who have implemented or are in the process of implementing 10646
without the need for additional profiling.
As 10646, just like other ISO standards, allows a lot of options,
that's simply impossile.
Second, Ohta-san's views are his own and do not represent those of the
Japanese committee (JSC2) that represents Japan at JTC1/SC2/WG2.
No, of course, though my view is, IMHO, common in the Internet in Japan.
The best way
for anyone to work to improve 10646 is to work within the standards process,
Don't mind. The Internet has its own standards process.
Finally, the name ISO 10646 should not be used to represent a subset of the
ISO 10646 standard, unless it is one of the subsets explicitly set forth in
the body of the standard. Therefore, the proposal should not use charset
names
based on the name ISO 10646 since they are in fact such subsets.
Section 13.1 of 10646 allows such character-wise subsetting, which makes
the entire standard pretty useless.
Such subsets
are, by definition, not in compliance with ISO/IEC 10646 and may not use its
name.
Section 13.1 says, to identiy such subset, list of all the names of
included graphic characters must be given. But as I quite honestry tried
to make the resulting charset name good, I didn't make the charset name
as the concatination of several thousands of character names such as:
charset=iso-10646-cjk-unified-ideograph-4e00-cjk-unified-ideograph-4e01-cjk-niied-ideograph-4e02-cjk-unfied-ideograph-4e03-cjk-unified-ideograph-4e04.....
The proposed subset names have not been proposed and thus not approved by
the ISO technical working group.
Do you like the blessed subset name:
charset=iso-10646-cjk-unified-ideograph-4e00-cjk-unified-ideograph-4e01-cjk-niied-ideograph-4e02-cjk-unfied-ideograph-4e03-cjk-unified-ideograph-4e04.....
?
In addition, they are misleading and can
cause confusion amongst a variety of implementors. Naming of subsets must be
proposed first then reviewed and finally approved and published by
JTC1/SC2/WG2.
Have you read RFC 1468? "ISO-2022-JP" is not the name registered to ISO.
We have IANA.
The ISO technical working group (JTC1/SC2/WG2) is very much interested in
seeing use of ISO/IEC 10646 on the Internet, and we hope through a
constructive dialog to establish the best way of doing so.
The best way to use 10646 on the Internet is to use it with OSI only.
Masataka Ohta