- From: Graham Klyne <GK@NineByNine.org>
- Date: Fri, 24 May 2002 11:23:52 +0100
- To: Dave Reynolds <der@hplb.hpl.hp.com>
- Cc: Dan Brickley <danbri@w3.org>, "Seaborne, Andy" <Andy_Seaborne@hplb.hpl.hp.com>, www-rdf-interest@w3.org
At 10:06 AM 5/24/02 +0100, Dave Reynolds wrote: >This is particularly significant when doing remote query because it not only >multiplies up the server time but it multiplies up the data that needs to be >transferred. Aha! I think I see where you're coming from here. I think that in the end one still want to be able to extract (some of) the variable bindings, but that could be done as a purely local (re)query on the returned graph? (I have already recognized the combinatorial issue as a potential problem with my approach, but for slightly different reasons. In some cases, in the simple application I developed, I really do want the software to return multiple combinations of variable bindings. But there's a fine line between getting enough and getting too much. I've addressed this by splitting up the query into a number of sub-queries that are joined-up by application logic - that's a rather ad-hoc solution.) #g ------------------- Graham Klyne <GK@NineByNine.org>
Received on Friday, 24 May 2002 06:53:28 UTC