The charset issue is very old. I started to participate in its
discussion in 1997, but it started away back in history. In this
message, I cover the history of XML media types and do not cover that
of HTML media types.
In 1996, IETF had a workshop for charset issues. RFC 2130 is a report
of this workshop, and RFC 2277 is based on it. RFC 2130 has the
following recommendation:
The recommended specification scheme is the MIME "charset"
specification, using the IANA "charset" specifications.
W3C I18N WG is also recommending the use of the charset parameter and
is trying to encourage WWW server developers to support the charset
parameter.
http://www.w3.org/International/O-HTTP-charset.html
http://www.w3.org/International/O-charset.html
After long discussion at W3C XML SIG, RFC 2376 was published. I am
forwarding some of the summaries Kurt and I wrote together. I believe
that then-discussion covers all points that have been raised recently
in this mailing list.
The I18N WG requested to the XML Syntax WG that determination of the
charset as described in RFC 2130 should not be changed, and the XML
Syntax WG said that there are no plans to change it.
http://www.w3.org/International/Group/1998/12/NOTE-i18n-rev-xml-19981221
http://lists.w3.org/Archives/Member/w3c-i18n-ig/1999Apr/0006.html
Although no new arguments have appeared, new WWW servers have
appeared. IIS 4.0 and 5.0 supports the charset parameter.
(I am forwarding some info written in Japanese to Martin.) In
my understanding, W3C I18N WG is encouraging such WWW servers.
I think we should just encourage conformant implementations without
changing the specification. Since new WWW servers provide better
support, we only have to improve them and use them correctly. The
advent of XML and Unicode provides a perfect chance.
The charset issue has made people burn out. I believe that
implementation and experiments are more important than repeating
the same agrument.
Cheers,
----
MURATA Makoto muraw3c(_at_)attglobal(_dot_)net