RE: [xml-dev] Re: determining ID-ness in XML

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Yes that is the case.  SOAP refusing to acknowledge it 
is a W3C problem.  They don't have the authority or perhaps 
the will to make their own specifications interoperable 
and well, so much for their standards.

The PI solution is the least impact solution so far.  Where 
one comes down on this seems to depend on how one thinks 
the system or processor vocabulary should be extended and 
what an XML processor must enforce.  Again, means have always 
been there and by the circular logic that seems to pervade 
the W3C thinking, people are free to ignore those means. 
So we add PIs.  They can ignore those too.  So we add 
more system vocabulary attributes; they can ignore those 
too.  Will we need Congress to legislate this stuff too?

The bizarreness of this to me Chris, is that these are 
well known issues of hypermedia regardless of the notation.

len

-----Original Message-----
From: Christopher R. Maden [mailto:crism@maden.org]

At 06:41 1-11-2001, Bullard, Claude L (Len) wrote:
>What would result from the reverse solution:  a document which
>is subject to a raw XPointer must have a DTD or Schema?

As Daniel noted, that's already the case.  The problem, however, is that 
even if I have a DTD, there is no guarantee that the receiving system is 
going to pay any attention to it, and so I don't know if my XPointers are 
going to work or not.  That's scary.

XML processors are required to acknowledge ID declarations in the internal 
subset[*], but that has been declared infeasible because SOAP (for whatever 
reason) forbids internal subsets.  It's also a maintenance pain since the 
ID-ness of an attribute is really a feature of the document type, and 
properly belongs with the rest of the document type definition, but some 
over-zealous validation systems will issue warnings about the duplicate 
declarations.  (For a while, MSXML would halt because an element type can 
only have one attribute of type ID - never mind that the two attributes had 
the same name, as well.)

I like the PI solution.  It's redundant with the DTD information, doesn't 
change the structure or naming of my documents at all, but can pass 
information on to systems that don't read the DTD without my having to muck 
about with an internal subset.