Re: Implementing statement grouping, contexts, quads and scopes

[pat hayes

>
> >But to be able to do anything worthwhile with the statements we need
> >to know the context of their assertion - they are asserted in this email.
>
> Well, that is a very controversial claim. WHY do we need to know the
> provenance in order to do anything worthwhile? What is wrong with
> having as a general assumption that one can usually forget about
> provenances (or maybe check them once and then forget about them) and
> just draw conclusions?
>

RDF in essence is a flexible database.  Triples in RDF are closely analgous
to rows in a database table (or many tables), except that they aren't really
keyed in any uniform way.  The only sure key is the compound key formed by
the three parts of a triple.

The original design of RDF works pretty well for this database-like purpose.
It supports database-like operations, like querying, intersections, etc.
(although even those go beyond the plain RDF).  Inferences you could make
from row data, you could make from RDF triples.

If you were to go to some data provider on the web and get data from its
database, you would probably have made a prior decision to trust it.  The
provider, we would hope. would have applied some QC.  Of course we know that
in practice, the data might still contain errors, but whether we would do
any QC ourselves on the data would depend on its importance, our history
with the provider, and so on.

Anything beyond this that we might want to do - I mean beyond retrieving
rows and collecting or combining the results - is new capability.  Colored
rows (I mean triples), sandboxes, more complex logic, etc.  It's good when
we can find ways to use the original design for these other things, but it
is not a surprise when we cannot.

To me, if we cannot arrive at a plausible and convenient way to to these
extensions within the bounds of basic RDF, there is no harm in developing
extensions.  The only thing is whether the extensions are agreed-on idioms
within the existing RDF syntax, like using reifications with some agreed-on
semantics, special terms like range and domain in RDFS, or of syntactical
extensions.

> >
> >Putting this more strongly, we only really have two contexts - 'in the
wild'
> >and 'in application X'.
>
> No, I think it will have to get much more delicate than that. For
> example, I might want to trust any source that is warranted as
> trustworthy by an ontology that I trust, or is owned by the US
> government. Others might have very different criteria, of course. In
> fact I forsee a whole economy of trustworthiness starting to emerge,
> eventually.
>

I agree with Pat here.  In fact, with PKI bridge authorities this kind of
thing is already happening, in a coarse-grained way and outside of RDF.

Cheers,

Tom P

Received on Thursday, 27 June 2002 18:02:25 UTC