Re: Querying reified statements

I'm stretching beyond what I really know here, but this sounds as if it 
might follow or map to the "dataguide" approach of Lore 
(<http://www-db.stanford.edu/lore/>)...

I could imagine a dataguide (or several) associated with a context.

#g
--


At 09:57 AM 12/13/00 -0800, Seth Russell wrote:
>David Allsopp wrote:
>
> > Perhaps the solution is to 'flatten' an RDF model containing reified
> > statements (applying filters according to the origin of the statements
> > or other criteria) to generate a model without any reifications, which
> > can be queried easily.  This could be done explicitly, or we could have
> > a Virtual Model which acts as a filtered interface to one or more actual
> > models?
>
>Yes I think you are on to something here.
>
>Whatever RDF is, I think it is safe to say, it is an external language, it
>is not very useful as your internal knowledge representation.
>
>So that to simplify the query I would use a quadruple internally instead of
>the external triple ... principally as Graham has suggested ... and place
>all statements made by someone in some context:
>
>[Context:Things Said About John]
>          |
>[Context:Sam says about John]----contains--->[id1, John, age,42]
>[Context:Tom says about John]----contains--->[id2, John, age,43]
>[Context:John says about John]----contains--->[id3, John, age,39]
>
>So if the context node [Context:Things Said About John] is selected, then
>all the triples id1, id2, id3 show up otherwise they do not;  so your
>simplified query would be something like:
>
>select [Context:Things Said About John]
>select [*, John, age, ?]
>
>The world is context sensitive, so things just work better in context :))
>Seth Russell

------------
Graham Klyne
(GK@ACM.ORG)

Received on Monday, 1 January 2001 14:01:20 UTC