[sc34wg3] TMQL, State of Affairs

[Murray Altheim]

Jan Algermissen wrote:
> On May 24, 2005, at 8:01 PM, Martin Bryan wrote:
>
>>explain why TMDM is so radically different from every other
>>data model in the universe
> 
> 
> Dunno what the significance of 'radically' is here. When data models  
> are different
> they are different, eh? So the QL of data model A does not work for  
> data model B.

I'm not sure why there is any confusion about this. As a
corollary example, as stated in the RDF Primer, "the
following documents contribute to the specification of RDF:"

   RDF Concepts and Abstract Syntax [RDF-CONCEPTS]
   RDF/XML Syntax Specification [RDF-SYNTAX]
   RDF Vocabulary Description Language 1.0: RDF Schema [RDF-VOCABULARY]
   RDF Semantics [RDF-SEMANTICS]
   RDF Test Cases [RDF-TESTS]
   RDF Primer (this document)

This makes the case that RDF, while like XTM having an XML
serialization syntax, has a distinct data model and model
theory. Query languages such as XQuery, suited to XML, while
able to query at an XML syntax level, cannot query at the
RDF expression level (much less at the RDF Schema level).
Likewise, an RDF query language couldn't query at the OWL
level.

So I don't see how XQuery is relevant here. As XML syntax,
XTM is a tree structure. XQuery is neither able to query its
graph structure (which is not apparent to generic XML
applications) nor the structures of its Topic Map data model/
model theory, which is yet another level up from its basic
graph structure.

I'm not sure if I've been exactly clear here, but I'm
confused by there being any confusion whatsoever about XQuery
not being suitable for querying XTM *as a Topic Map*. Yes,
one can obtain XML syntax-level structures, but Topic Maps
do not operate at the level of XML syntax-level structures.
Even a query language aware of XLink (which XQuery is not)
would still miss most of the Topic Map features. [I consider
it a major failing of the W3C's architectural principles --
or lack thereof -- that XQuery isn't natively XLink-aware.
Within the entire W3C corpus there's no way to query graph
structures, except simple tuples in RDF, and I'm not even
sure how one does that...]

I can certainly understand if this is simply a call to make
sure that these issues are addressed in the standard (which
is ostensibly what this sounds like). OTOH, if this is a
measure of confusion within the Topic Map community five
years after publication of the standard, I'm concerned.

Murray, hoping this doesn't add to the confusion

......................................................................
Murray Altheim                          http://www.altheim.com/murray/
Strategic and Services Development
The Open University Library
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

   In a May 4 interview with the CBS program "60 Minutes," to be
   broadcast on Sunday, Mr. Putin also reiterated complaints that
   America should not be lecturing him about rollbacks on democracy
   when "four years ago your presidential election was decided by
   the court."                        -- New York Times, 7 May 2005.
   http://www.nytimes.com/2005/05/07/international/europe/07prexy.html