Re: Extending the baseline with "asserted" stuff - syntax & semantics

Am 31. Juli 2024 10:18:45 MESZ schrieb Franconi Enrico <franconi@inf.unibz.it>:
>Following the long discussion on “asserted” stuff, some time ago I proposed to Thomas a way to extend the baseline capturing the gist of the discussion. Since I did not get any feedback on that,  

I made it public last Tuesday already, almost immediatly after I got it from you, in the proposal that I drafted at your request, and explicitly gave you credit for it [0]. For someone who seems to read my mails rather cursory you're making pretty strong statements. 

> in the last Semantics TF meeting I made the proposal public. This propoal should capture the gist of what has been discussed by Thomas, Souri, Gregory Williams, and others.
>
>The proposal is at https://github.com/w3c/rdf-star-wg/wiki/Extending-the-baseline-with-%22asserted%22-stuff.
>I paste it below for simplicity.
>
>Extending the baseline with "asserted" stuff (DRAFT 2024.07.31)
>
><https://github.com/w3c/rdf-star-wg/wiki/Extending-the-baseline-with-%22asserted%22-stuff#extending-the-baseline-with-asserted-stuff-draft-20240731>
>
>We extend the syntax and semantics of the baseline<https://github.com/w3c/rdf-star-wg/wiki/RDF-star-%22working-baseline%22> by adding a new property rdf:asserts, which has been called also rdf:states.
>
>
>Along this line, somebody

Is this now considered the appropriate way to refer to each other? It was me who proposed that, in the above mentioned proposal, and I see no reason why that shouldn't be said explicitly. At the very least it makes it easier to check prior conversations, ask for clarification, etc. And IMHO it just makes for a nicer environment. 

> proposes to change the name of rdf:reifies to rdf:describes, but by now we assume not to change the name in the baseline; a discussion about names in this extension may be necessary.
>
>
>We extend the well-formed syntax as follows:
>
>
>tripleStructure  ::=  ... | ( reifier rdf:asserts tripleTerm )
>
>
>
>and extend the interpretation of a triple if the predicate is rdf:asserts, in order to enforce the truth of the triple associated to the triple term as follows:
>
>
>[I+A](t) = TRUE   iff   <[I+A](t.s), [I+A](t.o)> ∈ IEXT([I+A](t.p)) and
>                        [I+A](t.o) = TRUE    if   t.p = rdf:asserts
>
>
>
>OBSERVATIONS
>
>
>:a rdf:asserts <<(:s :p :o)>> .
>  entails
>:s :p :o .
>
>:a rdf:asserts <<(:s :p :o)>> .
>  is equivalent to
>:a rdf:asserts <<(:s :p :o)>> .
>:s :p :o .
>
>
>
>This extension may be more suitable to extend RDFS and not RDF, since reasoning is needed to complete the graph.

Is the "macro" solution discussed in the past not also applicable here? 

>It has been suggested that the mapping between RDF-star and LPGs<https://github.com/w3c/rdf-star-wg/wiki/RDF-star-and-LPGs> should use rdf:asserts instead rdf:reifies.

My proposal goes further in several ways - in particular I argue that instantiation rather than reification should be considered the norm not just for RDF/LPG integration but for all of RDF - but this is indeed the most crucial piece. Well condensed! 


Best, 
Thomas 


[0] https://lists.w3.org/Archives/Public/public-rdf-star-wg/2024Jul/0109.html

Received on Wednesday, 31 July 2024 09:29:06 UTC