Re: Summary of the options on the table

On 25/01/2024 23:07, David Booth wrote:
> On 1/25/24 12:13, Pierre-Antoine Champin wrote:
>> . . .
>> https://htmlpreview.github.io/?https://github.com/w3c/rdf-star-wg/blob/main/docs/seeking-consensus-2024-01.html 
>>
>>    (source: 
>> https://github.com/w3c/rdf-star-wg/blob/main/docs/seeking-consensus-2024-01.md)
>
> Fantastic summary!   But I am surprised that I do not see any option 
> to harmonize with named graphs.  That seems like an obvious option to 
> consider, since :e (in your examples) is merely a fourth component 
> tacked onto the triple, and that's exactly what named graphs provide 
> -- see NQuads.
>
> Were named graphs already considered for RDF-star implementation and 
> rejected?  If so, why?

The general opinion is that this is inappropriate for a number of users, 
who use named graphs for different purposes (classing triples by 
provenance, or by permission, etc...). If adding quoted triples in a 
graph was creating "spurious" named graphs, this would potentially 
break, or at least interfere with, their business logic.

Also, named graphs are not covered by RDF semantics (for exactly the 
same reason: there are so many different uses of named graphs out there 
that any fixed semantics would not be consistent with some of them). Our 
ambition is to provide a standard (if minimal) semantics for RDF-star 
quoted triples.

Hope this helps.

> Please forgive me if this is a naive question from a distant observer.
>
> Thanks,
> David Booth
>