Re: (Lost in the noise perhaps - so asking again) - Is a trailing slash 'better' than a trailing hash for vocabs namespace IRIs?

Hi Christian,


On Tue, Oct 11, 2022 at 3:46 PM Christian Chiarcos <
christian.chiarcos@web.de> wrote:

>
> [PMcB] Well, as I stated in my response to Pierre-Antoine, I consider it a
>> fundamental principle of Linked Data that you can never fully know the
>> 'intended use case' for any shared vocabulary - as literally anyone should
>> be able to reuse that vocab, even within the confines of a closed
>> Enterprise (e.g., via mergers or acquisitions in the future).
>>
>
> True, but the creator of a vocabulary typically does. For vocabularies
> whose usage of slash or hash really presents a problem, natural selection
> will kick in in the end. But I actually think this is a minor concern.
>
>

[PMcB] Yeah, I absolutely agree - "For vocabularies whose usage of slash or
hash really presents a problem..."  - yep, then clearly you have viable
grounds for breaking with Best Practice, so off you go (and use hash!).



>   the other Best Practice of providing `rdfs:isDefinedBy` triples per
>> term.
>>
>
> I actually have a bit of an issue with this as a best practice, because it
> doesn't have to refer to the full RDF vocabulary. It may be used for this
> purpose, but in fact, I've been frequently using it to link concepts with
> the URIs of the *unstructured documents* that define them. And these are
> unstructured because there just is no normative RDF vocabulary for that
> particular domain. Think of language identifiers. ISO 639-1 and -2 are
> available as RDF, but ISO 639-3 is not, so the normative reference for any
> language not covered by ISO 639-1 or -2 is a bunch of human-readable tables
> on a website, and if I wanted to provide the "is defined by" context of,
> say my:EgyptianArabic, that would not be the my:-ontology, but
> https://iso639-3.sil.org/code/arz. I don't think there is another RDF(S)
> property to serve that purpose (not within the RDF[S] namespace, but we
> could go with dc:source, but that has somewhat different semantics;
> similarly, owl:sameAs would be wrong because it comes without an RDF
> formalization, and we cannot really be sure whether
> https://iso639-3.sil.org/code/arz is meant to be a class or an instance,
> whereas my:EgyptianArabic if it is proper OWL2/DL, should be one or the
> other).
>
>
[PMcB] Interesting! I wouldn't really mind which predicate becomes the Best
Practice - I just like 'rdfs:isDefinedBy' because it feels so intuitive,
and it exists in one of the 'core' RDF vocabs already (and it seems to be
the most commonly used in practice - not that that's a particularly strong
justification!)

But I'd still suggest that your use of 'rdfs:isDefinedBy' to point to the
URIs of *unstructured documents* might not be strictly correct. I mean,
here's the actual complete definition of that term (complete with typo in
the value of the 'rdfs:comment' triple :) !):

  rdfs:isDefinedBy a rdf:Property;
    rdfs:isDefinedBy <http://www.w3.org/2000/01/rdf-schema#>;
    rdfs:label "isDefinedBy";
    rdfs:comment "The defininition of the subject resource.";
    rdfs:range rdfs:Resource;
    rdfs:domain rdfs:Resource;
    rdfs:subPropertyOf rdfs:seeAlso.

So that is clearly communicating that the 'rdfs:range' should be RDF. So
yeah, I probably would have suggested 'dcterms:source' too (which has no
`rdfs:range`), or prehaps 'prov_o:wasDerivedFrom' (but that has the range
'prov_o:Entity').

I'm not sure why you suggest 'dcterms:source' has different or
inappropriate semantics for you, as it's 'dcterms:description' seems to fit
your use case pretty well to me - i.e., "This property is intended to be
used with non-literal values. The described resource may be derived from
the related resource in whole or in part. Best practice is to identify the
related resource by means of a URI or a string conforming to a formal
identification system.".

Anyways, I'd certainly be open to suggestions for a different/better
predicate to use as Best Practice guidance for linking a vocab term back to
its defining/containing/describing vocab. But that would be a separate
issue/thread though!



> If we make *that* a best practice recommendation, we should also propose
> another property for the other use case (pointing to the defining resource
> if it is not an RDF vocabulary).
>
>
>> If a vocabulary provider relies massively on OWL2 inferences to
>>> pre-populate the ontology with, say, static values for properties of
>>> individuals of a particular type, then these will not be found (or need to
>>> be explicitly searched for) if only the local URI and its immediate context
>>> is returned. In other words, defaulting to shlashes might be a possible
>>> strategy for ontologies *only if they don't make use of implicit triples*.
>>>
>>
>> [PMcB] I might be missing something here, but would simply dereferencing
>> the vocab namespace IRI not solve this issue?
>>
>
> Yes.
>
>
>> In other words, if this community came to agreement on the Best Practices
>> of [use-slashes + provide-`rdfs:isDefinedBy`-triple-per-term +
>> dereferencing-vocab-namespace-IRI-gives-full-vocab-metadata], then vocabs
>> that followed that Best Practice wouldn't have any problem here - client's
>> could just dereference the namespace IRI and get everything they need in a
>> single server repsonse, right?
>>
>
> Yes -- if we agree on that best practice ... not 100% sure we will, mostly
> because the semantics of rdfs:isDefinedBy are actually broader than this
> use case. This is one use case it covers, and it is mentioned as such
> ("may") in the spec, but it is not the only one.
>
>
[PMcB] Yep, I agree. As above, I'd certainly be open to suggestions (but I
still feel 'rdfs:isDefinedBy' is 'fine' too!), but yeah, discussion for a
separate thread...


> The alternative thing to do would be to create/resort to a designated
> property to provide the full vocab. That could be dct:isPartOf ("A related
> resource in which the described resource is physically or logically
> included.") Wrt. a best practice for pointing to the containing vocabulary,
> I would prefer that over rdfs:isDefinedBy, and it's already an established
> property from an established vocabulary.
>
>
[PMcB] Yep, 'dcterms:isPartOf' definitely seems a worthy contender! As
above, discussion for a separate thread I think...


> But I'm all for a best practice recommendation of that kind for slash URIs.
>
> Best,
> Christian
>

-- 
This e-mail, and any attachments thereto, is intended only for use by the 
addressee(s) named herein and may contain legally privileged, confidential 
and/or proprietary information. If you are not the intended recipient of 
this e-mail (or the person responsible for delivering this document to the 
intended recipient), please do not disseminate, distribute, print or copy 
this e-mail, or any attachment thereto. If you have received this e-mail in 
error, please respond to the individual sending the message, and 
permanently delete the email.