Re: Request for feedback about RDF-star

Kjetil, thanks for your feedback;

my comments below

On 25/03/2021 00:17, Kjetil Kjernsmo wrote:
> onsdag 24. mars 2021 10.20.44 CET skrev Pierre-Antoine Champin:
>> We aim to mint a new
>> IRI to be used with RDF-star. In your opinion, is it
>> acceptable/desirable to propose its inclusion in the RDF namespace [3],
>> or should we instead mint it in a separate namespace?
> Good question! Here's my personal take:
>
> First of all, I note that there exists a TAG finding on a similar question:
> https://www.w3.org/2001/tag/doc/namespaceState.html
> It encourages the authors of the namespace document to declare how they
> expect the namespace to evolve. It does not seem that the RDF namespace
> document does this, which is understandable, since the original RDF syntax
> recommendation predates that TAG finding.
>
> However, as Ora seems to be indicating the RDF Syntax Spec says:
> https://www.w3.org/TR/rdf-syntax-grammar/#section-Namespace
>
> "Any other names are not defined and SHOULD generate a warning when
> encountered, but should otherwise behave normally.", which seems to
> indicate that the expectation is that the names is a fixed set.
>
> I recall that the we went through some pains as the SKOS namespace changed
> in the last moments of the spec work, so there is some precedent for that
> too. I think it all went fine even though I had to change it in code that
> were in customer's hands.
>
> I would tend to agree that the new name seems to belong in the rdf:
> namespace, I would be in favor of it being added there eventually.
>
> Still the question remains as to what a CG should do, it could do the
> guerrilla tactics and use the rdf: namespace from the start (I wouldn't be
> offended), but given the RDF Syntax Rec's rather clear expectation, I think
> it is appropriate to wait.
>
> My suggestion is to mint a URI outside of the rdf: namespace, but make it
> clear in any notes or WDs that it is expected that it will change.

As I hinted in my original email, we know from experience that this does 
not work, unfortunately.
Once an IRI gets used "out there", changing it in the spec becomes a 
breaking change for too many people, and so it does not happen.

Otherwise,

* the RDF1.1 would have migrated all the rdf: and rdfs: names into a 
single, shorter, namespace;

* the FOAF IRI would have changed its 0.1 for a 1.0 (or no versioning at 
all);

* the content-type application/x-www-form-urlencoded would have been 
registered without the x- prefix;

* ...

   best

>   Then, I
> think that the issue should be settled by a WG shortly before the
> publication of the Candidate Recommendation, as the CR is also a
> solicitation of implementations beyond the WG. In the interest of URI
> Persistence, it is appropriate that it is fixed by the time the world beyond
> the WG is asked to take note.
>
> Cheers,
>
> Kjetil
>
>
>