Re: New abstract for RDF Concepts

Much better would be:

RDF datasets comprise a default graph and zero or more named graphs.   The graph names, which are IRIs, can be used in RDF statements.


On 11/08/2012 08:32 AM, Richard Cyganiak wrote:
> Peter,
>
> On 8 Nov 2012, at 06:45, Peter F. Patel-Schneider wrote:
>>> RDF datasets comprise a default graph and zero or more named graphs, and are used to to express metadata about RDF graphs,
>> How?
> By making statements about the graph name.
What does this have to do with the graph itself?
>
>>> and to organize data by context.
>> How?

Ditto.
> By putting data with different context in different named graphs.
>
>> As far as I can tell just about the only thing that the WG should say about RDF datasets is that they consist of a default graph and zero or more named graphs, and can be used to associate names with graphs.
> What is wrong with making factual statements about what they are actually used for?

In the primer, it might be acceptable to be so sloppy, but not, I think, in 
Concepts.
>
>> If the WG calls out two questionable uses of RDF datasets, then it should also call out all the other, less-questionable uses. Who is volunteering to enumerate all of these?
> Well, how about you make a start by naming one?

provenance, merging, sourcing, for starters.
>
> Best,
> Richard

peter