Re: "Multi-Edge Support in RDFn" slides

On 16 Dec 2022, at 17:29, Pierre-Antoine Champin <pierre-antoine@w3.org> wrote:

I think the whole paradox thing is a distraction. I started it, and I shouldn't have -- my initial concern was about the ability to make syntactically self referential statements, paradoxical or not.
RDF's semantics is to weak to express paradoxes, and so will (probably) by RDF-star's.
Paradoxes may occur in semantic extensions, but we are not there yet. Let's find consensus of the basic semantics first.

I fully agree.
RDF, RDFS, OWL, are all unable to express paradoxes.
They can, however, express knowledge which does not make any sense in the domain you want to model.
That’s why we need to be explicit about best practices on how to use RDF as a modelling language, whenever it is possible.
The reification “macro” has some semantic assumptions that make it meaningful; we have to spell them out explicitly.
cheers
—e.