Subject: Re: Preparing for XPointer - msg#00023

At 03:49 AM 29/10/01 +0100, Chris Lilley wrote:
>> During that thought experiment, it occurred to me that we should
>> be watching +xml registrations to ensure that, where possible,
>> "#foo" syntax is reserved for identifying fragments labelled with
>> an ID (or http://www.w3.org/2001/XMLSchema#id) type attribute of
>> value "foo".
>
>Yes.
>
>And other schems for example #xpty(foo) or #xpath(foo) or #bar(foo)
>would be clearly distinguishable.
>
>> This would ensure that XPointer could be be retrofitted into
>> 3023 later and apply to all */xml and */*+xml types.

Erk. Danger, Will Robinson. This is in danger of tripping over
what is maybe the #1 gaping architectural hole as regards XML & the
Web. The problem is that at the moment, given some arbitrary XML,
there is *no* good way to determine what's an ID without recourse
to some external resource like a DTD or schema, and that, to use
a technical term, sucks.

One could dismiss the problem and say "well, XHTML defines what
an ID is, so does SVG, and so on, and so there's no problem in
the general case because there's no general case". But I'd like
to have foo#bar be meaningful even if "foo" is served up as
application/xml.

Since fortunately there isn't much in the way of deployment of
hyperlinks trying to point into XML documents, it is NOT TOO LATE
to bite the bullet and do something radical to fix this. I at
least will be arguing in favor of this whenever I get a chance.
Which if successful would probably have an impact on one or more
media-type RFCs. -Tim