- From: Graham Klyne <GK@ninebynine.org>
- Date: Thu, 25 Aug 2011 13:47:32 +0100
- To: Luc Moreau <L.Moreau@ecs.soton.ac.uk>
- CC: public-prov-wg@w3.org
On 22/08/2011 22:52, Luc Moreau wrote: > Hi Graham, > > I read the proposed changes, and the document offer two different > solutions for the cases where there is a uri for provenance or there is not. Indeed, this was a requirement. > Text of section 2 is not as balanced I feel. > > It starts with "A general expectation is that ... access provenance information > ... dereferencing its URI", > which seems at odd with ... "If there is no URI associated ..." in the third > paragraph. If the problem is one of "balance", I see no need for such here. This is for the web, and according to web architecture there is a clear preference for associating URIs with resources: http://www.w3.org/TR/2004/REC-webarch-20041215/#pr-use-uris I think it's entirely appropriate that we should *encourage* developers to allocate and use URIs for accessing provenance. #g -- > It would be better to be upfront, and say that there are two cases, case 1: > where there is a URI, case 2, where > there is no URI. > Luc > > On 29/07/11 09:07, Provenance Working Group Issue Tracker wrote: >> PROV-ISSUE-55 (are-provenance-uris-needed): Are provenance URIs really needed >> [Accessing and Querying Provenance] >> >> http://www.w3.org/2011/prov/track/issues/55 >> >> Raised by: Luc Moreau >> On product: Accessing and Querying Provenance >> >> >> I would like to initiate a debate about a fundamental assumption of the PAQ >> document: "A general expectation is that web applications may access >> provenance information in the same way as any web resource, by dereferencing >> its URI.". >> >> I can see that this "expectation" may be valid in a number of circumstances. >> But in various projects, we have implemented provenance stores as stand-alone >> services, accumulating provenance about things. Whenever the provenance of >> something was requested, we were querying the storage system, and returning >> the set of assertions that was appropriate. >> >> The use of a provenance-uri is counter-intuitive in this context. I would even >> argue it puts an undue burden on the provenance store. Indeed, the provenance >> store would have to maintain a reverse mapping provenance-uri to thing-uri, so >> that the query about that thing can be re-issued, if required. (Of course, see >> ISSUE-54 on the requirements set on provenance-uris and what they refer to.) >> >> What do people think? >> >> >> >> >
Received on Thursday, 25 August 2011 13:52:02 UTC