Skip
repetitive navigational links
L-Soft  -  Home of  the  LISTSERV  mailing list  manager LISTSERV(R) 14.5
Skip repetitive navigational links
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (October 2004)Back to main ZNG pageJoin or leave ZNGReplyPost a new messageSearchProportional fontNon-proportional fontLog in
Date:         Tue, 12 Oct 2004 15:49:20 +0100
Reply-To:     "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Sender:       "Z39.50 Next-Generation Initiative" <[log in to unmask]>
From:         Mike Taylor <[log in to unmask]>
Subject:      Re: CQL modifier ordering (Pay attention Ralph:))
Comments: To: [log in to unmask]
Comments: cc: [log in to unmask]
In-Reply-To:  <[log in to unmask]>
              (message from Robert Sanderson on Tue, 12 Oct 2004 14:07:02 +0100
              (BST))

> Date: Tue, 12 Oct 2004 14:07:02 +0100 (BST) > From: Robert Sanderson <[log in to unmask]> > > > Then when given a query like > > title any/fb.broader/foo.resolveURI "http://a.b.c/d.f" > > a well-behaved server would respond with an error: > > 501 Request is stupid. Make a sensible one instead. > > It might also blindly look in its thesaurus for 'http://a.b.c/d.f' and not > find anything, and then resolve the URI. Right. Not because it falls back to resolving the URI as a special case when looking in the thesaurus first fails, but because the thesaurus lookup succeeds, saying that the set of all terms equivalent to "http://a.b.c/d.f" is { "http://a.b.c/d.f" }. So it blindly goes on and resolves all the URIs in that set. Right? > For example, if the terms are unique identifiers which happen to be > URIs (eg searching a repository of identifiers) then the server > can't really tell if that's the right order. True. The moral here seems to be (once again) that we should not impose arbitrary restrictions on domain-specific communities, because they will surely have problems that we've not thought of yet. > So... my point... if the server can't process the order then it > should give a diagnostic certainly. But it shouldn't make value > judgements about whether an interpretation is meaningful or not. I agree. > eg: title any/phonetic/stem "cwms" [1] > > [1] My favourite word for hangman and scrabble :> Cym fjord bank glyphs vext quiz. _/|_ _______________________________________________________________ /o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk )_v__/\ A day without sauropods is like a day without sunlight. -- Listen to free demos of soundtrack music for film, TV and radio http://www.pipedreaming.org.uk/soundtrack/


Back to: Top of message | Previous page | Main ZNG page

LISTSERV.LOC.GOV CataList email list search Powered by LISTSERV email list manager