Re: Decision from the Semantics TF: liberal baseline

Hi Thomas

> On Jan 6, 2025, at 6:06 PM, Thomas Lörtsch <tl@rat.io> wrote:
> 
>> On 2. Jan 2025, at 17:29, Franconi Enrico <franconi@inf.unibz.it> wrote:
>> 
>> At the last Semantics TF we discussed about the RDF semantics of the liberal baseline. 
>> In the current version of the document:
>> https://github.com/w3c/rdf-star-wg/wiki/RDF-star-%22liberal-baseline%22#rdf-semantics
>> the two discussed restrictions
> 
> 
> Shouldn’t we rather stop calling them restrictions? They define things, but they can’t enforce anything.

E.g., the second proposal says that objects of rdf:reifies should have type rdf:proposition. It is true that, if an object resource does not have that type stated, we don't assume it does not have the type (negation as failure, which is contradicted by the open-world assumption); rather, we assume that the resource has that type, by inferring that type for it. But, it is still a restriction IMO - objects need to have type rdf:proposition - even though it is not being "enforced" as in throwing an error or something. 

> 
>> of RDF semantics are formalised:
>>    • triple terms, appearing in triples or in triple terms, are of type rdf:proposition;
> 
> That seems redundant: triple terms are always of type proposition, they can’t be anything else, or can they? I.e. they _are_ propositions. To me it seems like it would be enough for them to get their own class, in analogy to rdfs:Literals.
> 
> I remember the consensus in the SemTF discussion leaning towards the rdf: namespace, but when I look at <https://www.w3.org/TR/2014/REC-rdf-schema-20140225/#ch_classes> I find the rdfs: namespace more fitting, putting rdfs:Proposition right besides rdfs:Resource, rdfs:Class and rdfs:Literal.
> 
>>    • objects of the property rdf:reifies, appearing in triples or in triple terms, are of type rdf:proposition.
> 
> 
> Mildly in favor of it, as an expression of intent. The property rdf:reifies has a very specific purpose, and expressing that axiomatically might help prevent misuse.
> 
> However, looking at the RDFS entailment patterns in <https://www.w3.org/TR/rdf11-mt/#patterns-of-rdfs-entailment-informative>, 
> some of them seem rather verbose, like
> 
> rdfs4a 
> xxx aaa yyy . 
> ->
> xxx rdf:type rdfs:Resource .
> 
> some seem useful, like 
> 
> rdfs11 
> xxx rdfs:subClassOf yyy .
> yyy rdfs:subClassOf zzz . 
> ->
> xxx rdfs:subClassOf zzz .
> 
> But they all seem evidently valid. The proposed reif axioms however aren’t, they can easily be broken. That rather speaks against them.
> 
>> We may decide to have both restrictions, any of them, or none.
>> Let’s open the general discussion :-)
> 
> Glad to see that you still seem to enjoy it :-)
> 
> .t
> 
> 
>> —e.
>> 
>>> On 13 Dec 2024, at 18:20, Franconi Enrico <franconi@inf.unibz.it> wrote:
>>> 
>>> Today the Semantics TF met, and we agree to submit to the working group a proposal for a liberal baseline. It is summarised in <https://github.com/w3c/rdf-star-wg/wiki/RDF-star-%22liberal-baseline%22>, to be discussed (and voted?) at the first focussed meeting in 2025.
>>> Basically, there will be a no syntactic restriction in using both rdf:reifies and triple terms. 
>>> Reification is sanctioned only if it makes use of the property rdf:reifies or any of its subproperties; the subject of rdf:reifies is called a reifier.
>>> Triple terms would be always of type rdf:Proposition, and the range of rdf:reifies would be rdf:Proposition.
> 
>

Received on Tuesday, 7 January 2025 00:18:50 UTC