Proposed XML contexts [was Re: Proposed URI classes]

/ Paul Grosso <> was heard to say:
| in the business of classifying URIs.  We are identifying
| useful XML contexts for external identifiers. 

Yes. Sorry. That's what I meant. Bad, norm! No donut!

| Anything that does not have a defined function and context in
| XML 1.0+namespaces (think the Infoset) makes me uneasy for xmlcat
| version 1.0, and that includes include.  (I consider the stylesheet
| PI to be of the same generation as XML 1.0+namespaces, but XInclude
| isn't even a CR yet, and other forms of include require application
| dependent knowledge even to recognize them.)

I'm not talking about XInclude per se. It seems to me that
we could put xsl:include, xsl:import, xsd:include, xsd:import, and
XInclude in one bucket, but we don't have to.

I left out 

schemalocation

And I guess the appropriate context to use for resolving
xsl:include/imports is "stylesheet" and for xsd:include/import it's
schemalocation. And we just leave XInclude out of the mix this time
around.

                                        Be seeing you,
                                          norm

-- 
 | The best people usually owe their
XML Technology Center     | excellence to a combination of qualities
Sun Microsystems, Inc.    | which might have been supposed
                          | incompatible.--Bertrand Russell