Re: first cut at proposing closure for the RDFCore legacy issue list

On Apr 15, 2011, at 14:03 , Pierre-Antoine Champin wrote:

> On 04/14/2011 03:18 PM, Dan Brickley wrote:
>> rdf-equivalent-uri's: Should RDF have a mechanism for declaring two
>> uri's to be equivalent?
>> 
>> RESOLVE: Closed, owl provides owl:sameAs already.
> 
> Unless I was really the only one to argue in favour of including it in
> core RDF, I think that matter is still under discussion.

Do you mean that we would duplicate the owl:sameAs predicate in the RDF vocabulary? And what about the other, similar predicates? Eg, same predicate, same class? After all, sameAs is only for instances (although often misused for the equivalence of classes). 

I am a bit concerned that it may not be THAT easy, and not really necessary. OWL has it, there is no problem in using it in RDF data and no, you do not need a complex OWL DL reasoner to handle that properly; OWL RL has it (and its friends) which can be handled with a few rules...

> 
>> rdfms-literalsubjects: Should the subjects of RDF statements be
>> allowed to be literals
>> 
>> CONTINUE: the situation is unclear. In a sense, literals are
>> resources. Restrictions are largely (but not entirely) syntactic.
> 
> my understanding is that the restriction is in the abstract syntax (and
> hence in the concrete syntaxes), but not at all in the semantics.

This was the subject of a HUGE discussion; for many, literals as subject is a big NO-NO from a modeling point of view. The early discussion leading to the charter showed a deep division on this subject, certainly no consensus; hence it is explicitly listed as out of scope in the charter:

http://www.w3.org/2011/01/rdf-wg-charter#outofscope


> 
>> rdf-bnode-predicates: Request to allow b-nodes as property labels
>> 
>> CONTINUE: is the abstract syntax / formal semantics already happy with
>> this? Does it affect ntriples, turtle etc?
> 
> the abstract syntax exclude it
> http://www.w3.org/TR/rdf-concepts/#section-Graph-syntax
> 
> my impression is that the semantics does not even care, but I may miss
> subtle details
> 
> I guess ntriples and turtle could be straightforwardly extended to
> support that. RDF/XML would probably be left behind, though...
> 
> I can see nice use cases for allowing that, but I already ear voices
> whispering "not in the charter"... :-)

Exactly. See pointer above :-)


> 
>> rdfms-contexts: Suggestion that the concept of context is missing from RDF.
>> rdfs-layered-subset: A request to define subset of RDFS with a more
>> conventional layered architecture
>> 
>> RESOLVE: CONTINUE to bear this in mind as Semantics are revised /
>> improved based on deployment experience.
> 
> I don't know if it is specified anywhere (writing the mail in the plane,
> so I can't check rihght now), but I recall reading mentions to RDFS-LD,
> a restriction of RDFS compatible with the layered model of description
> logics.

I am not sure what context is. JSON-LD has the term context but I am afraid it is for something else: it is the JSON object that contains prefix mapping, base setting, stuff like that...

Ivan


> 
> nice job Dan
> 
>  pa
> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf