[> > = quote from MURATA Makoto]
Gavin Thomas Nicol writes:
<?xml-stylesheet type="text/xml" href="#style1"?>
We have agreed that fragments do not have media types. What will
happen to such media types combined with fragment identifiers?
Should they be simply ignored?
This is another case where the fragment inherits the media type.
Again, the entire document must be fetched (as text/xml), and then
the fragment identifier resplved in terms of that. With the example
above, the fragment shoould be resolved in the resource containing the
reference.
I'm not sure. Makoto poses a very interesting question. Either this
example is an exception to the rule, or there is something wrong with
(this use of) the xml-stylesheet PI.
In this example, there is no sense in putting the type of the document
in the PI, because we must have already established the type of the
document in order to understand the PI.
What we don't know yet, is what type the style sheet is. If it is
embedded, it is probably CSS or XSL, but how do we find out which?
Heuristics will work in this case, of course, but that is not a
scalable solution. It seems to me that we have no other place to put
the type of the fragment than on the PI...
Bert
--
Bert Bos ( W 3 C ) http://www.w3.org/
http://www.w3.org/people/bos/ W3C/INRIA
bert(_at_)w3(_dot_)org 2004 Rt des Lucioles / BP
93
+33 (0)4 92 38 76 92 06902 Sophia Antipolis Cedex, France