Re: Keep mode explici [Re: Annotation syntax]

On 7/09/2020 16:09, Pierre-Antoine Champin wrote:
> On 05/09/2020 20:34, Olaf Hartig wrote:
>> On lördag 5 september 2020 kl. 11:37:09 CEST Ghislain Atemezing wrote:
>>>> Le 4 sept. 2020 à 00:30, Holger Knublauch<holger@topquadrant.com>  a écrit
>>>> [...]
>>>> If RDF* in the most general sense is SA mode then the PG mode could be,
>>>> for example, called RDF+ aka RDF plus. It would be a bit like OWL Full vs
>>>> OWL DL, or SHACL-SPARQL vs SHACL Core. Some tools will elect to support
>>>> PG mode/RDF+ only.
>>> +1. I like this analogy. Probably RDF+ can confused some of us using RDFS+
>>> (as a profile to do reasoning). What about saying RDF* when you support
>>> both SA and PG (like OWL Full), then RDF*-XX (XX = SA or PG) if someone
>>> supports just one of them?
>> +1
>
> I think that an even better way to /not/ confuse people would be to 
> get rid of modes altogether! To quote Pat Hayes on this list in a 
> previous message [1]:
>
> > these are not two ‘modes’ but two languages, indistinguishable in 
> syntax but with different semantics. That is a truly terrible idea.
>
> One benefit that I saw in the discussion about "annotation syntax" was 
> to propose a different syntax for the "PG-mode use cases", which would 
> have allowed us to keep the original syntax (<< >>) exclusively for 
> the "SA-mode use cases" -- but maybe I was misinterpreting.
>
> So if really there is a consensus that << >> should be interpretable 
> in two different ways depending on the "mode",
>
No, there is no consensus for that. << ... >> would always represent a 
triple node that may be subject or object of an RDF* triple, and this 
triple node exists regardless of whether the actual triple is also 
asserted in the graph or not. The additional syntax would only add an 
optional shortcut notation for two statements at once, see previous 
emails. Definitely no different "modes" here, but SA mode allows more 
graphs than PG mode.

Holger


> then let me suggest the following:
>
> add to RDF* and SPARQL* a directive @mode (or maybe MODE for 
> SPARQL*...), akin to @prefix or @base, to make the mode explicit. So 
> one would either write:
>
>   @mode PG.
>
>   <<:bob :worksFor :ACME>> :since 2018.
>
> or
>
>   @mode SA.
>
>   :alice :believes <<:bob :worksFor :ACME>>.
>
> If @mode was absent, I would prefer the default value to be SA, but I 
> understand that some implementations already assume PG mode, so maybe 
> the default value could be left unspecified for backward compatibility.
>
>   pa
>
>
> [1] https://lists.w3.org/Archives/Public/public-rdf-star/2019Sep/0052.html
>
>