- From: Steve Harris <steve.harris@garlik.com>
- Date: Tue, 21 Jul 2009 09:42:08 +0100
- To: Simon K Johnston <skjohn@us.ibm.com>
- Cc: W3C SPARQL Working Group <public-rdf-dawg@w3.org>, public-rdf-dawg-request@w3.org
On 21 Jul 2009, at 01:13, Simon K Johnston wrote:
> As mentioned in the minutes I had mentioned last week that in trying
> to work through the details of MINUS and UNSAID I wonder if one way
> to get ourselves unstuck is to look at and document some use cases.
> We do already have NOT EXISTS in SQL (and XQuery) and these are well
> known and used today so one might ask what they are used for, what
> queries require these capabilities? For example we might be able to
> come up with a set of simple queries, expressed in natural language,
> that we can then evaluate the two proposals against.
>
> For example, "find me all records where no due date has been set"
> seems to map well to UNSAID, but seems more clumsy when expressed
> with MINUS - or rather the natural expression seems harder to read
> in the resulting SPARQL.
So, I'm not sure I follow this. Probably I'm missing some subtlety in
"find me all records where no due date has been set", but taking the
exact keyword out of the equation for a minute I would expect the
expression to be:
SELECT ?rec
WHERE {
?rec a :Record .
UNSAID {
?rec :due-date ?date
}
}
in either case.
- Steve
Received on Tuesday, 21 July 2009 08:42:49 UTC