- From: Chris Bizer <bizer@gmx.de>
- Date: Mon, 23 Feb 2004 15:48:14 +0100 (MET)
- To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- Cc: www-archive@w3.org
Hi Jeremy,
a side comment and not our main topic now, but I have been thinking a little
bit more about the query language. Maybe I should change the syntax and move
the query patterns from statement level to graph level and closer to the
TriG syntax.
Example: Select all persons which have been borne before 1970, using only
statements from authors I trust.
SELECT ?y
WHERE
?x (?y rdf:type ex:Person .
?y ex:birthyear ?z)
null (?x saidby ?a)
null (ex:Me trusts ?a)
AND ?z < "1970"
and allow the abbreviation of graph names - as you proposed - if they don't
matter, like in the following version of the query
SELECT ?y
WHERE
?x (?y rdf:type ex:Person .
?y ex:birthyear ?z)
(?x saidby ?a)
(ex:Me trusts ?a)
AND ?z < "1970"
Both only syntactical changes but the queries look more like named graphs
:-)
Chris
--
GMX ProMail (250 MB Mailbox, 50 FreeSMS, Virenschutz, 2,99 EUR/Monat...)
jetzt 3 Monate GRATIS + 3x DER SPIEGEL +++ http://www.gmx.net/derspiegel +++
Received on Monday, 23 February 2004 09:51:29 UTC