Issue #1 : Semantic neutrality of openguid identifiers

[bernard]

Seems to me openguid semantics should be as void as possible, and in
particular not interfere with semantics of linked vocabularies.
Looking at http://openguid.net/specification#ontologies
Seems a good heuristic to infer oguid:identical statements from
owl:sameAs, owl:equivalentClass or skos:exactMatch and the like. But
the other round, like Example 5, seems dangerous to me, even if the
resources have the same rdf:type, like owl:Class. Openguid identity is
really at the level of identity of referent, not identity of
representation. In the case of Example 3, suppose I have

<SampleOne> owl:equivalentClass <SampleTwo>
<SampleOne> oguid:identical <SampleThree>
<SampleThree> rdf:type owl:Class

The declared semantics of classes <SampleTwo> and <SampleThree> may
well be inconsistent, because those two classes represent two views of
the same concept. So inferring their equivalence is not a good idea.

In general, I think Openguid statements could be inferred from
statements in other languages, but not the other way round. Openguid
semantics is too weak for that. And its strength comes from this very
weakness and absence of meaning (in a very taoist view of the world).

bernard

[ja...@openguid.net]

After much deliberation, I have to say I agree with you on this. Does
that make me taoist?

Though it is tempting to say much about the world, it is not
rightfully the domain of this service. That would be better left to
things like UMBEL.

I would like to keep inferring identical semantics from
equivalentClasses, etc. This allows semantic web processing engines
to expand their associative abilities.

Thanks for the input, I will remove this implication in the next
iteration of the spec.

On Sep 25, 6:44 am, bernard <bernard.vat...@mondeca.com> wrote:
> Seems to me openguid semantics should be as void as possible, and in
> particular not interfere with semantics of linked vocabularies.

> Looking athttp://openguid.net/specification#ontologies