Pete said:
> > > > To do this, I'd need to store along with the
> > > > assertions some explicit indication of the fact that
> the specific
> > > > assertions that form my set/record are related (e.g.
> came from a
> > > > common source RDF/XML doc on the Web) - I think in the
> RDF context
> > > > this is described as "context" or "provenance".
Sigge said:
> Is that the same as what used to be called a reification?
> Wouldn't a record RDF in that case be the set of assertions
> that are asserted by the same creator at the same time. OK,
> that is an operational definition, but anyway.
I guess I was side-stepping slightly exactly what the criteria for
"record-ness" might be ;-) But your operational definition sounds a good
starting point! I was just saying that there may well be a relation
between a set of assertions which it might be important to retain in the
context of the "assertion store".
If there was a simple one-to-one correspondence between "record" and
"source RDF/XML document", then I guess the assertion store application
could rely on recording that against each triple/assertion, but as soon
as that simple correspondence breaks down (e.g. multiple "records" in an
RDF/XML source doc) or you are passing sets of aggregated assertions
from one store to another, then, yes, it seems to me that you need some
_explicit_ metadata saying that assertion 1 and assertion 2 are part of
a set/"record". I've never used RDF reification in practice (and I
suspect that there are subtleties I haven't grasped until I try!), but
my reading of it is that it might be an appropriate mechanism for
recording that metadata.
Pete
|