[MOBY-dev] RDF descriptions of MOBY services

[Phillip Lord]

>>>>> "Mark" == Mark Wilkinson <mwilkinson at mrl.ubc.ca> writes:

  Mark> Hi all,

  Mark> Nina and I are debating the most appropriate way to represent
  Mark> a MOBY service signature in RDF.  The main source of
  Mark> controversy is with the inputs and outputs, in particular when
  Mark> it comes to collections.

  Mark> In the current draft, I have the input as a bnode type:Bag,
  Mark> with individual input articles coming off of it.  Input
  Mark> articles in MOBY are either Simple or Collection, however at
  Mark> the moment I have made this distinction implicit - if the
  Mark> article is type:Bag, then it must be a Collection, and if it
  Mark> is not type:Bag, then it represents a Simple, and has various
  Mark> information (object_type, namespace, etc.) predicated to it.

  Mark> What concerns me is this: If we make the interpretation
  Mark> implicit, then we don't need additional predicates, whereas if
  Mark> we explicitly define the edge to an article as "has_simple",
  Mark> or "has_collection", then these predicates need to be defined.
  Mark> The data structure (Bag) is semantically identical (IMO) to
  Mark> what a Collection article is in MOBY, so I don't see a *need*
  Mark> to explicitly define this... but under either circumstance the
  Mark> person designing a query or a parser would need to understand
  Mark> how to interpret either the "shape" of the graph (if we leave
  Mark> it implicit), or the meanings of the predicates (if we make it
  Mark> explicit).

  Mark> ...??  This never ceases to befuddle me.  It seems that, even
  Mark> with RDF, we can not get away from the requirement of
  Mark> community agreement on interpretation.  I guess this reminds
  Mark> me of Phil's recurring statement "ontologies are only useful
  Mark> if they are shared"...

  Mark> ...but from those of you with more experience in RDF than I
  Mark> have, could you advise me which of these two options is "more
  Mark> correct"?  Is it better to be explicit, and define new
  Mark> project-specific predicates, or is it better to let the data
  Mark> structure implicitly speak for itself using existing RDF/S
  Mark> predicates?



Mark

I'm not convinced that it makes any difference. Either way you have to
have an interpretation of the RDF wrt to the application at
hand. This is the same with all uses of RDF; the RDF serialisation of
OWL-DL, for example, does not tell you how to interpret the semantics
of OWL-DL. You have to read the OWL spec for that.

Cheers

Phil