 |
 |
Thomas - I take your point about compete vs. incomplete description with respect to "RDF representation", but I'm not using it in that context, but rather in the context of: if you dereference a web resource, you get a representation of that resource; there may be many representations, one of which may be RDF. NOT to imply that that representation is complete. Clearly there is confusion over "resource" - web resource, information resource, RDF resource. I'm not sure it's useful to sort through it all, as some of these definition have historical significance only. So let's try to break this down without talking about resources. An RDF property may be a datatype property or an object property. A datatype property takes a literal for its object (Quotes). An object property takes an "individual" for its object (braces). "Individual" as in, a member of an RDF class. An individual always has an RDF description (or in your terminology, you can always make an RDF statement about it). Do you agree? Ray > -----Original Message----- > From: Bibliographic Framework Transition Initiative Forum > [mailto:[log in to unmask]] On Behalf Of Thomas Berger > Sent: Tuesday, November 17, 2015 4:08 AM > To: [log in to unmask] > Subject: Re: [BIBFRAME] Properties of Item proposal > > Am 16.11.2015 um 22:14 schrieb Denenberg, Ray: > > Karen, take this one step at a time. Suppose I have a triple: > > > > <some subject> bf:someProperty <http://someURI> > > > > Where again I am using turtle notation and I am explicitly enclosing > > http://someURI in angle braces. I claim that http://someURI > > (enclosed in angle braces) must be dereferenceable as RDF (meaning > > that there is an RDF representation of that resource, http://someURI, > > which may be one of many representations, and I can get that RDF > > representation by doing a get on it requesting RDF via content > > negotiation). > > No. You may be alluding to the concept of "information resources" when > distinctions between Real World Object and Web Content describing them > comes into play. But these may be anything and are not restricted to RDF. > > I would say, already thinking in lines of "an RDF representation of that > resource" is alien to RDF: Having <http://someURI> in object position of a > statement implies that there is an entity which can be identified by that IRI. > So somewhere in the world there may exists statemtents with > <http://someURI> in subject position, or there may not. That the sum of > these statements yields a (complete?) description of that entity or even that > there is (at least) some single place where a description in the sense of > several RDF triples for that IRI exist, is not implied. If so, we would have the > case of an information resource again, but stating that there is an associated > information resource on the web for <http://someURI> would mandate the > coining of a different IRI since at that point we have to differentiate between > the resource and its description as another resource. There is a standard > practice with fragment identifiers to achieve that without effort. > > The hassle with IRIs (<bracketed>, as identifiers in RDF context) against URIs > as strings in "quotes" is based on our attempts to encompass the concept of > "Real World Identifiers" (a notion probably originally koyned here): Since we > are expecting to "see" that https://viaf.org/viaf/174144647695570462777/ > "is" a VIAF identifier (or a canonical URI for the VIAF identifier > "174144647695570462777" or an alternative representation for the abstract > VIAF identifier of VIAF identifier datatype behind that) or that "urn:isbn:0- > 486-27557-4" refers to a resource which also will be unanimously identified > by the ISBN 0-486-27557-4. For that we need /inspection/ of the URIs and > that is due to opaqueness not possible for the bracketed Identifiers in the > RDF sense. > > viele Gruesse > Thomas Berger