RE: (proposal) was Re: defn of Named Graph

On Friday, October 04, 2013 4:54 AM, Pat Hayes wrote:
> On Oct 3, 2013, at 10:01 AM, Markus Lanthaler wrote:
> > ... as I've just re-read the definition of RDF source in Concepts, it
> > appears that its definition is wrong (or at least inconsistent):
> >
> >   We informally use the term RDF source to refer to a persistent yet
mutable
> >   source or container of RDF graphs. An RDF source is a resource that
may be
> >   said to have a state that can change over time. A snapshot of the
state can
> >   be expressed as an RDF graph
> >
> > So a RDF source may return *multiple* RDF graphs, i.e., a dataset
> 
> No, it can return different graphs at different times, but it always
> returns a single graph.

I understand what it should say but in my opinion it doesn't (clearly
enough).


> The dataset equivalent would be a thing like a
> dataset but where the inner graphs were RDF sources, so that we would
> have 'named sources' rather than named graphs. Which is exactly what we
> need to make the idea of 'naming' to work properly. There is currently
> no terminology for this dataset-source notion, so we will have to
> invent one or adapt a current term to this wider use.

Exactly. That's why I proposed defining something like a "RDF graph
source"...


--
Markus Lanthaler
@markuslanthaler

Received on Friday, 4 October 2013 10:08:54 UTC