Re: Minutes of DAWG 2005-02-15 telcon for review

On Tue, 2005-02-15 at 16:27 +0000, Dave Beckett wrote:
> Minutes of RDF DAWG telecon 2005-02-15 for review

Thanks for getting these out promptly.

A correction...

> Discussion of constraints expressed as triples vs them in AND clause.
> Some things may not be (so easily) expressed as triples compared to
> them in the expression language.  Experience here from n3.  The
> examples in 0127 had literals as subjects so weren't good for sparql
> as defined now.  Alberto presented some  ical/RDF-cal examples using
> contraints into triple-patterns (e.g. dates comparinsons) at
>   http://demo.asemantics.com/zparqler/examples/example8.html  
> 
> 	PREFIX ical: <http://www.w3.org/2002/12/cal/ical#>
> 	SELECT ?uri ?summary ?dtstart
> 	FROM <rdfstore://localhost:1234/xannouncements>
> 	WHERE 
> 	    (?uri ical:summary ?summary)
> 	    (?uri ical:dtstart ?dtstart)
> 	    (?dtstart op:dateTime-greater-than "2000-06-00T00:00:00A")
> 
> Two problems here - literal subjects and "2000-06-00T00:00:00A" isn't
> a xsd:dateTime literal.
> 
> Discussion of the set of predicates that sparql engines must
> understand - the op:* above.  EricP reports that the current rq23 draft
> does not require recognising this.  Alberto and DaveB remembered this
> being present once but it was removed by AndyS [for reasons the
> scribe lost track of in the discussion - interoperability?].
> 
> ACTION JosD: Make 2 test cases over the same data using the
> op:dateTime predicate and without using it.

I think it was the other way around:

ACTION JosD: make 2 test cases using the same query but with
different data: one without an op:dateTime triple in the input,
and one with.


-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Tuesday, 15 February 2005 18:29:32 UTC