Re: Significant W3C Confusion over Namespace Meaning and Policy

/ David Orchard <dorchard@bea.com> was heard to say:
| Is there in effect a substitution rule being applied to xml:id in xml:id
| unaware XML 1.0 processors?  Seems like the substitution rule is roughly
| to treat xml:* attributes as regular attributes, modulo the extra
| complications.

I don't know about a substitution rule, from a parser perspective, xml:*
attributes *are* just regular attributes. An application might or might
not know what xml:lang means, an application might or might not know
what xml:space means, an application might or might not know what xml:base
means, an application might or might not know what xml:id means, and an
application might or might not know what xml:foobar means.

If we consider, as many specs do, that the Infoset is a description of
what a process requires on input and/or produces on output then in
every case, when processing is layered (application on top of xinclude
processor on top of parser, etc.), different layers in the stack might
or might not change the way the infoset appears to subsequent layers
in the stack.

In particular, an XInclude processor might expand XIncludes. An
xml:base processor might change the [base uri] property, and an xml:id
processor might change the [attribute type].

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM / XML Standards Architect / Sun Microsystems, Inc.
NOTICE: This email message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.