- From: James Cheney <jcheney@inf.ed.ac.uk>
- Date: Fri, 27 Jan 2012 12:25:26 +0000
- To: Paul Groth <p.t.groth@vu.nl>
- Cc: "public-prov-wg@w3.org" <public-prov-wg@w3.org>
The semantics currently has two levels, but allows entities (and other objects) to have attributes that change over time. I think the two choices are to have two levels: objects (entities, agents, activities, events, with attributes that don't change over time) records (syntactic descriptions of things). or to have three levels: things (with attributes that change over time) objects (entities, agents, activities, events - with attributes that don't change over time) records (syntactic descriptions of things). The problem with the two-level approach is that it is not clear at the semantic level how we should relate different entities that "speak about the same thing", since there are no things. One solution (which IIRC Graham and Stian suggested) is to think of a thing as a collection of entities related by "alternateOf". But since we haven't really agreed what alternateOf is yet, this seems to put the syntactic cart before the semantic horse. By the way, I think that the discussion at the end of the teleconference yesterday was about whether we can or should consider records to be objects. --James On Jan 27, 2012, at 10:55 AM, Paul Groth wrote: > Hi James, > > In your semantics document you ask the question: > > - Should we have two levels (things/entities vs. records) or three levels (things vs. entities vs. records)? > > I just wanted to get clarity on your stance. I think it is that we have three levels but that for the purposes of the describing provenance we only discuss entities and records? Essentially, we are interested in describing the surrogates (i.e. entities) for "real world" objects. > > Is that correct? > > thanks, > Paul > > -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.
Received on Friday, 27 January 2012 12:26:06 UTC