notes from xml2003 discussion (paul's intro)

paul cotton:


b/g context...

w3c rdfq wg discussion at AC meeting.
"we should talk" from rdfq folk.
is there a desire for converge/collide or go off in diff directions.
talked in hallways... suggested a bof here; W3C Tech Plenary not till
March, so meeting here.

Some rdf and xml folks here, others couldn't make it...

rough agenda from paulc (see slides)

current status.

id of special rdf query and data model requirements.

peter f -- mentioned work in europe on xquery as a high level ql 
over top of other qls

(18 attendees)

AGENDA:
1. current status xquery/xpath
2. current status rdf and ql
3. id of special rdf query and data model reqs
4. id of pros and cons of using xquery/xpath
5. future acions, discussions

Michael Rys:
re (2) does this mention market issues, acceptance etc.
does www have resources/desire for yet another ql?


attendees -- see paper circulated by jon robie



http://www.idealliance.org/papers/xml2001/papers/html/03-01-04.html


---

PaulC

XQuery WG
....longest without a REC
....started out as largest
....big spec. 

went to LC on 12th.

[admin: agreed to make proceedings of meeting public]

LC ends mid feb. 3 month LC.
Want to make sure get good review from RDF and OWL folks. To get their 
view on whether we did good job of making a QL for XML.
Our method: first we did a data model for XML.
We don't necc insist that the datamodel is generated from 'physical'
xml.

the data model doc tells you how you'd go about creating the datamodel
instance.
incl schema input etc.
the QL operates thru a closed set of operators over that datamodel.
one objection we've heard: that you cant use xquery over rdf as rdf's 
syntax is overflexible. But... maybe its the mapping of rdf to xquery
datamodel that is the key.

xquery 1.0 is a superset of xpath 2.0 (readonly path expressions)
xquery allows construction, new instances, bunch of other stuff
...we have a separate data model doc.

[...]

Received on Wednesday, 10 December 2003 19:53:59 UTC