- From: Miles, Simon <simon.miles@kcl.ac.uk>
- Date: Sun, 25 Mar 2012 14:15:09 +0100
- To: "public-prov-wg@w3.org" <public-prov-wg@w3.org>
Hello Graham, I think the alternate/specialisation concepts are even more important given the facility for people to use the concepts in a lightweight/looser way. To adapt Stian's recent example, Graham is a journalist contributing to the BBC site, while Simon is a blogger quoting from the BBC site. They do not know each other or coordinate. Graham's PROV data refers to the precise version of an article with a date-specific URI, as he might edit the article later: wasQuotedFrom (:bbcNews/article254/2012-03-25, http://en.wikipedia.org/wiki/Provenance) Simon's PROV data refers to the article in general: wasQuotedFrom (:my-blog-2012-03-26, http://www.bbc.co.uk/news/article254/) Without someone, possibly Graham, also expressing some relation between the two article entities, e.g. specializationOf(:bbcNews/article254/2012-03-25, http://www.bbc.co.uk/news/article254/) then we have no hope of extracting the possible connection between :my-blog-2012-03-26 and http://en.wikipedia.org/wiki/Provenance from the mass of PROV data. Wouldn't this be a real, and even common, use case? Wherever two people make assertions about the same thing, either they have to coordinate or a relation between the references has to be drawn separately. While I'd agree these relations are mostly not particular to provenance, alternate and specialisation seem to be a helpful minimum to include in PROV. And specialisation, at least, seems directly relevant to provenance, as what happened to the specialisation also happened to the generalisation at some time. thanks, Simon Dr Simon Miles Senior Lecturer, Department of Informatics Kings College London, WC2R 2LS, UK +44 (0)20 7848 1166 Using Normative Markov Decision Processes for Evaluating Electronic Contracts: http://eprints.dcs.kcl.ac.uk/1398/ ________________________________________ From: Graham Klyne [GK@ninebynine.org] Sent: 25 March 2012 08:18 To: Luc Moreau Cc: public-prov-wg@w3.org Subject: Re: PROV-ISSUE-29 (mutual-iVP-of): can two bobs be mutually "IVP of" each other [Conceptual Model] In my review comments which I think you have yet to get round to, I question whether we actually need to have these concepts in the DM. Originally, by my recollection, they were introduced to explain the relationship between provenance entities and (possibly dynamic) real world things. With the looser description of the provenance model terms, I don't see why this level of detail is needed in the data model. The purpose of separating the DM description from its sytriucter interpretation is, IMO, to provide a relatively easy point of entry, particularly for developers who might generate provenance information for artifacts created by their software. It may be that, when dealing with the stricter interpretation of provenance per Part 2, these ideas need to be re-introduced. But in that context, they could be introduced in purely formal terms, as additional constraints that underpin some of the other constraints. Then, I think such descriptions could be rooted in the formalization (per semantics) of the notion of invariance of entity provenance. #g -- On 23/03/2012 15:00, Luc Moreau wrote: > Dear all, > > This is the oldest issue in the tracker for prov-dm. > > A week ahead of the release of prov-dm wd5 (for internal review), I propose > to close this issue. > > Please have a look at the section [1] currently describing > alternateOf/specializationOf. > As indicated yesterday, suggested english definition for these relations is > appreciated. > > Regards, > Luc > > > [1] http://dvcs.w3.org/hg/prov/raw-file/default/model/prov-dm.html#component4 > > > On 07/11/2011 12:22 PM, Provenance Working Group Issue Tracker wrote: >> PROV-ISSUE-29 (mutual-iVP-of): can two bobs be mutually "IVP of" each other >> [Conceptual Model] >> >> http://www.w3.org/2011/prov/track/issues/29 >> >> Raised by: Stephen Cresswell >> On product: Conceptual Model >> >> >> As it currently stands, I believe that it does not exclude the possibility >> that two bobs may be mutually "IVP of" each other - >> i.e. you could have bobs A, B such that (B IVPof A)& (A IVPof B), and this is >> surely not intended. >> >> This could arise if, for bobs A, B : >> - A and B both represent the same entity >> - A and B share some immutable properties, and they have corresponding values. >> - B has some immutable properties which correspond to mutable properties of A >> - A has some immutable properties which correspond to mutable properties of B >> >> Possibly the asserter-defined test (included in "IPV of" definition) that real >> world states modelled by A and B are "consistent" may disallow >> "IPV of" in this situation. However, unless that is guaranteed, I think that >> the definition of "B IPV of A" (if it is still to have a definition) should >> additionally require that: >> "A has no immutable properties which correspond to mutable properties of B" >> >> Stephen >> >> >> >> >
Received on Sunday, 25 March 2012 13:15:44 UTC