RE: SKOS Consistency

Antoine/Alistair,

Thanks for your answers.

I have one more question wrt the integrity of SKOS dataset. The SKOS
Reference states that skos:prefLabel, skos:altLabel and skos:hiddenLabel
are all disjoint. This leads inconsistencies when skos:prefLabel and
skos:altLabel and/or skos:hiddenLabel have the same value for a given
language. The reference also states that no concept can have 2
skos:prefLabel for the same language.

Suppose now that I have two SKOS concepts as follows:

<A>  skos:prefLabel "stringA"@en;
          skosxl:altLabel <C> .

<B> skos:prefLabel "stringB"@en;
         skosxl:altLabel <C> .

<C> rdf:type skosxl:Label ;
        skosxl:literalForm "stringC"@en .

Would this be consistent?

Kind regards,

Quentin

-----Original Message-----
From: Alistair Miles [mailto:alimanfoo@googlemail.com]
Sent: donderdag 8 juli 2010 12:53
To: Quentin Reul
Cc: public-esw-thes@w3.org
Subject: Re: SKOS Consistency

Hi Quentin,

On Tue, Jul 06, 2010 at 10:47:09AM +0200, Quentin Reul wrote:
> Hi all,
>
> I was looking at the SKOS reference [1] and I was unable to determine
> whether a SKOS model would be consistent if no skos:hasTopConcept
> property was defined within a concept scheme.

Just adding a general comment to Antoine's nice reply, the open world
assumption which is part of the RDF and OWL semantics means that an
absence of some data cannot give rise to inconsistency.

I.e., you can only become inconsistent by saying things (and contradicting
yourself), not by forgetting to say something.

Checking for "missing" data is, however, very useful in some
circumstances. E.g., if you have a file which you know is supposed to
contain all the data for a complete thesaurus, you might then check to see
if any concepts are missing a prefLabel in some language. In this case,
your check effectively assumes a closed world. This is a perfectly
reasonable thing to do, but this type of checking is beyond the scope of
the SKOS reference.

I.e., the SKOS reference *will* tell you if your data is inconsistent (in
an open world). It *will not* tell you if your data is missing anything,
because how you define missingness is up to you, and you will want to vary
your definition depending on what you're trying to achieve.

Hth,

Alistair

>
> Cheers,
>
> Quentin
>
> [1] http://www.w3.org/TR/skos-reference
>
> --
> Quentin Reul
>
> Semantic Technology Consultant
> TenForce BVBA
> Haachtsesteenweg 378
> 1910 Kampenhout
> Belgium

--
Alistair Miles
Centre for Genomics and Global Health <http://cggh.org> The Wellcome Trust
Centre for Human Genetics Roosevelt Drive Oxford
OX3 7BN
United Kingdom
Web: http://purl.org/net/aliman
Email: alimanfoo@gmail.com
Tel: +44 (0)1865 287669

Received on Thursday, 8 July 2010 12:54:58 UTC