- From: Thomas Lörtsch <tl@rat.io>
- Date: Mon, 25 Mar 2024 19:40:22 +0100
- To: "Lassila, Ora" <ora@amazon.com>
- Cc: Franconi Enrico <franconi@inf.unibz.it>, RDF-star Working Group <public-rdf-star-wg@w3.org>
In my understanding an reifier is a specialization of an identifier (it’s probably safe to say that refier is a subclass of identifier). It is an identifier used in the subject position of an rdf:refies relation. We need a way to refer to these subjects of rdf:reifies relations, and "reifier" might be a good enough name. Best, Thomas P.S., since the issue came up a few times recently: IMHO the http-range14 problem applies to an identifier used as reifier just as well as to any other identifier on the semantic web: only circumstances might (to be fair, mostly will) disambiguate if the identifier is used/intended to refer to the addresssed resource itself (e.g. a webpage, an email address, a triple term) or to what that resource denotes (its meaning). > On 25. Mar 2024, at 17:03, Lassila, Ora <ora@amazon.com> wrote: > > Well, it *looks* like an identifier, and it is *used* like an identifier, so now we are back to my general worry: how to explain. I am not happy with saying that there is something that is just like an identifier but is not an identifier. RDF is already confusing enough for many people. > If I say > :c :p1 “some value” > and then > << :b | :s :p :o >> :p2 “some other value” > I have effectively said > :b :p2 “some other value” > and will be hard pressed to explain that :c is an identifier (of a resource) but :b is not. I think this problem becomes even clearer (and the example more confusing) if I write it in a way where the IRIs look like typical IRIs: > <http://example.com/this/that#xyz> :p1 “some value” . > <http://example.com/this/that#zxy> :p2 “some other value” . > I am not happy. > Ora > From: Franconi Enrico <franconi@inf.unibz.it> > Date: Monday, March 25, 2024 at 8:06 AM > To: "Lassila, Ora" <ora@amazon.com> > Cc: RDF-star Working Group <public-rdf-star-wg@w3.org> > Subject: RE: [EXTERNAL] A single reifier can reify more than one triple term > CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe. > Lassila, Ora <ora@amazon.com> wrote: >> >> >> Aside from this, I also worry about having to explain an identifier reifying a set of triples vs. an identifier identifying a set of triples (a named graph). I promise that distinction will be lost on many people. > > First of all, let’s not call it identifier; reifier is, so far, a better and LESS CONFUSING name. > A reifier is a resource the is unique only in the case the triple term it reifies uniquely identifies it. Whenever the triple term does not uniquely identify a reifier (like in the liz and Richard marriage) then you allow multiple reifiers for the same triple term. > And for any such triple terms not uniquely identifying a reifier, NECESSARILY do exist other triple terms that form another incomplete way to identify a reifier. The union of all those triple terms would completely identify uniquely a reifier, but we can not express this in RDF, which can only express binary predicates. > —e.
Received on Monday, 25 March 2024 18:40:41 UTC