Re: dataset semantics

On 12/17/11 10:58 AM, Pat Hayes wrote:
> BTW, what was the rationale for having a nameless graph in a dataset in the first place? Seems to me that the SPARQL design would be improved if all graphs were required to have some kind of name, and the query was obliged to use the name. After all, this is how the rest of the Web works.
Yes, and that's the case in our implementation. If you want a default 
graph you explicitly designate a given named graph as such. That's it. 
Otherwise, you have queries scoped to all graphs or specific lists of 
named graphs in a FROM or FROM NAMED clause.

In our world view SPARQL named graphs are just named partitions that 
hold 3-tuple based record collections. Depending on use-case, we can 
make statements about named graphs using their IRIs.

Trying to marry the named graph world views of SPARQL and RDF is a 
mercurial pursuit. You also have the dbms/store world views to factor 
into the mix and that only makes it more mercurial to pin down. In our 
case, if more granularity is sought e.g., making statements about 
statements we do so courtesy of terms from reification oriented ontologies.



-- 

Regards,

Kingsley Idehen	
Founder&  CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen