Re: 6.3 -- proposal for (informal) dataset semantics

On Sat, 28 Apr 2012 10:25:04 -0500, Pat Hayes <phayes@ihmc.us> said:

    phayes> One niggle on terminology: there is no such thing as an
    phayes> "IRI-labelled node". The things in the RDF triples are
    phayes> just IRIs. (Your termoinology suggests a distinction
    phayes> between the 'node'and the IRI used to label it, and this
    phayes> is potentially confusing. Hence my bud-nipping.)

Something about that seems strange... Can't quite put my finger on
it. Something about the vertices related to others by certain
labelled edges and just happening to carry some IRI-labels. But given
what I understand you to have written about contexts and islands (or
neighbourhoods) the vertices are embedded in the context and the
IRI-labels can actually sanely be reused in other contexts on what
must really be different vertices. Right? Does that make any sense at
all?

    phayes> raises the issue of just how, if at all, they can ever get
    phayes> asserted. Do you have any idea in mind for specifying how
    phayes> I can use a name to assert the graph named by the name?

How about:

    :phayes rdf:asserts :somegraph.

It seems to me, hidden right inside your question ("how I can use"),
assertion is a speech act and so requires a speaker. Can it really be
so easy to write down?

Cheers,
-w
--
	    William Waites MBCS <wwaites@tardis.ed.ac.uk>
 Visiting Researcher, Laboratory for Foundations of Computer Science
	    School of Informatics, University of Edinburgh