Re: 3023 update (was Re: Agenda TAG Telcon: 8th Nov 2004)

* Tim Bray wrote:
>That's only if you serve it as text/xml or application/xml, which I 
>assume you don't.  I assume you serve it as something useful, for 
>example text/html or application/xhtml+xml (I'm writing this on an 
>airplane so can't check).

You seem to think there is a difference between serving an XHTML
document as application/xml or application/xhtml+xml, many people
seem to think there is no such difference and it seems many tools
do not care much. http://www.w3.org/MarkUp/2004/xhtml-faq#ie is a
good example here. I believe the claim that "XML-based browser, of
which at the time of writing means more than 95% of browsers in use,
can process new markup languages without having to be updated." is
based on the assumption that there is no such difference.

>In either of those cases, the language comes with a well-defined way to
>use fragment IDs.  This is precisely the reason I'm pushing back - it's
>only useful in the case where something is served as */xml, and when I
>observe that in the wild, it's usually an error (e.g. RSS feeds) which
>is cheerfully ignored by the receiving software anyhow.

You miss that this has an impact on general purpose XML processing,
fragment identifiers are highly relevant to XML Linking for which
an often cited use case is link recognition in unknown XML formats
for link checkers and search engines.

Received on Saturday, 20 November 2004 00:13:53 UTC