W3C home > Mailing lists > Public > ietf-charsets@w3.org > January to March 1999

Re: UCS PUA and charset tags

From: Ned Freed <Ned.Freed@INNOSOFT.COM>
Date: Mon, 08 Mar 1999 09:05:42 -0800 (PST)
To: Misha Wolf <misha.wolf@reuters.com>
Cc: ietf charsets <ietf-charsets@iana.org>
Message-id: <01J8KZ0P3FGO91WROK@INNOSOFT.COM>
> I recall seeing a mail (from Ned?) stating that any uses of the UCS
> Private Use Area (PUA) would require their own charset tags, but can't
> remember which list it was on.  I'm asking because MathML makes heavy
> use of the PUA and, presumably, uses the standard charset tags.

I don't recall making such a statement, but it seems obvious that some sort of
label would be necessary if you expect other systems to understand and respect
such usage.

> A follow-on question is:

> If use of the UCS PUA requires additional charset tags when the text
> is encoded using one of the UCS-based CESs (eg UTF-16), what happens if
> the text is encoded using a more basic charset (such as US-ASCII),
> coupled with rules provided by a higher-level protocol, eg XML's
> numeric character references?

It seems to me it would depend on the rules such references follow. If such
references can be qualified with an appropriate charset then there's no
problem. But if not, then you have a potential interoperability problem,
especially given that one of the goals of XML-derived formats is to provide for
interoperability between applications.

				Ned
Received on Monday, 8 March 1999 12:14:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 5 June 2006 15:10:50 GMT