- From: Thomas Lörtsch <tl@rat.io>
- Date: Thu, 15 Feb 2024 16:35:09 +0100
- To: Franconi Enrico <franconi@inf.unibz.it>
- Cc: "Peter F. Patel-Schneider" <pfpschneider@gmail.com>, "public-rdf-star-wg@w3.org" <public-rdf-star-wg@w3.org>
> On 15. Feb 2024, at 16:24, Franconi Enrico <franconi@inf.unibz.it> wrote: > > On 15 Feb 2024, at 16:11, Peter F. Patel-Schneider <pfpschneider@gmail.com> wrote: >> The only real problem I see with option 1 is that it is easy to create Franken-reifications. For example >> << :e | :s :p :o >> :x :y . >> << :e | :s1 :p1 :o1 >> :x :y . > > To me, this should be possible: > << :w3 | :bill-clinton :related-to :hillary-rodham >> :starts 1975 . > << :w3 | :1st-female-NY-senator :wife :42nd-potus >> :starts 1975 . > Indeed, well-formedness in option 2 allows for that. It would help to be more precise. I guess what you, Enrico, refer to is the issue of co-denotation. That should be possible as RDF standard reification is referentially transparent and talks about the meaning of a statement, not its syntactic form. I intuitively understood Peter as refering to two reifications that in common understanding don’t share a common meaning, e.g. << :e | :weather :is :good >> :x :y . << :e | :jeans :are :blue >> :x :y . Is this distinctin between co-denoting and not co-denoting reifications the non-trivial problem that Peter refers to? In any case it’s good to know that option 2 has the same problem ;-) And as option 3 is said to be semantically equivalent to option 2 we might just conclude that every proposal has to deal with this problem one way or the other... Thomas > PS: this is not an endorsement of option 1, which is my least favourite option. > > cheers > —e.
Received on Thursday, 15 February 2024 15:35:21 UTC