- From: Richard Cyganiak <richard@cyganiak.de>
- Date: Mon, 18 Oct 2010 22:23:31 +0100
- To: Eric Prud'hommeaux <eric@w3.org>
- Cc: Alexandre Bertails <bertails@w3.org>, Paul Tyson <phtyson@sbcglobal.net>, ashok.malhotra@oracle.com, Michael Hausenblas <michael.hausenblas@deri.org>, RDB2RDF WG <public-rdb2rdf-wg@w3.org>
On 18 Oct 2010, at 21:53, Eric Prud'hommeaux wrote: > How about either a > stronger predicate: > <http://foo.example/DB/People#_> rdfdb:hasAttribute People:ID . > or some type annotations: > <http://foo.example/DB/People#_> a rdfdb:Relation ; > rdfs:member People:ID . > People:ID a rdfdb:RelationAttribute . > > My preference is the former, possibly with some domain and range > assertions for rdfdb:hasAttribute, which would look like: > rdfdb:hasAttribute rdfs:domain rdfdb:Relation ; > rdfs:range rdfdb:RelationAttribute . > > Does that meet your requirements? Is it attractive enough? Either of these work for me. We should at some point consider standardizing our terminology. In R2RML I'm eager to use the SQL spec's terminology (which would be Table and Column, or possible BaseTable and Column). You use relational algebra terminology. I guess this is better discussed and decided later when we have more material on the table, so for now I'm ok with the above. Richard > > >> Richard > > -- > -ericP
Received on Monday, 18 October 2010 21:24:07 UTC