- From: Andreas Harth <andreas@harth.org>
- Date: Tue, 19 May 2009 21:53:39 +0100
- To: Hugh Glaser <hg@ecs.soton.ac.uk>
- CC: "semantic-web@w3.org" <semantic-web@w3.org>
Hugh Glaser wrote: > On 18/05/2009 19:36, "Andreas Harth" <andreas@harth.org> wrote >> I like the idea to put more intelligence into the web infrastructure. >> > Er, I don't. > The effectiveness of the web has always been that any "intelligence" is > provided by the servers, services and clients, and if we try to do otherwise > for the semantic web we will be left in a backwater. > I'd like to get some form of standardised query functionality instead of "plain" URI lookups. The ability to pose queries (maybe even only a subset of SPARQL, but with keyword search and some form of top-k processing) is a requirement for the type of retrieval systems I'd like to build, without relying on a centralised repository. I know the Harvest guys tried to build a distributed web system a while ago and their keyword search part didn't survive, but what's a keyword search compared to an arbitrarily complex SPARQL query. You seem to have bought into the whole SPARQL thing, so not sure if we really disagree... Regards, Andreas.
Received on Tuesday, 19 May 2009 20:54:22 UTC