Re: [ISSUE-77] [ISSUE-48] Re: [Dbpedia-discussion] Skos subject properties are deprecated

Leonard, and all

Leonard's point goes along the same line as my previous message. Let's 
extend the notion of document to be equivalent to "resource", and we're 
done. That was all the point of the URI specification to begin with. 
Another hit on that nail : The notion of document is extended in many 
ways in various communities to anything bearing information. The first 
information on a thing being to be made distinct from the continuum of 
the universe, through naming, identifying, and asserting distinctive 
properties (such as a URI), as long as something is identified, it bears 
information, hence can be considered a document.
A bit farfetched maybe, but closing the debate, conceptually and 
technically.

Bernard

PS : I vote for "A", of course.

Leonard Will a écrit :
> On Thu, 24 Jan 2008 at 13:24:12, Richard Cyganiak <richard@cyganiak.de>
> wrote
>   
>> All,
>>
>> I agree with Bernard that SKOS needs a property for attaching  resources to
>> concepts.
>>
>> The problem with skos:subject at the moment is this: The Core Guide
>> gives the impression that the domain of skos:subject is documents  only.
>> But there is no explicit domain declared in the vocabulary  definition.
>> Furthermore, several parties (e.g. DBpedia) have a clear  need for a property
>> that relates non-document resources to  skos:Concepts.
>>
>> I think there are three options for resolving this:
>>
>> A) Clarify that the domain of skos:subject is indeed any resource, and  that
>> the term “subject” is used loosely here.
>>
>> B) Clarify that the domain of skos:subject is documents only, and
>> introduce a new super-property of skos:subject that explicitly covers  any
>> resource. It could be named for example skos:category, or
>> skos:indexedAs.
>>
>> C) Clarify that the domain of skos:subject is documents only, and  leave the
>> task of defining a property non-document resources to others.
>>
>> My preference would be, in that order, B), A), C).
>>
>> Thanks,
>> Richard
>>     
>
> It may be of interest that in BS8723-1:2005 we have used the term
> "documents" as a label for any kind of resources. The relevant
> definition reads thus:
>
>     "document
>
>     item that can be classified or indexed in order that it may be
>     retrieved
>
>     NOTE  This definition refers not only to written and printed
>     materials in paper or microform versions (for example, books,
>     journals, diagrams, maps), but also to non-printed media,
>     machine-readable and digitized records, Internet and intranet
>     resources, films, sound recordings, people and organizations as
>     knowledge resources, buildings, sites, monuments, three-dimensional
>     objects or realia; and to collections of such items or parts of such
>     items."
>
> This is perhaps another illustration of the distinction between a
> concept and the term which has been chosen to label it. I think that the
> scope note above covers all the "non-document resources" that Richard
> refers to.
>
> Leonard Will
>
>   

-- 

*Bernard Vatant
*Knowledge Engineering
----------------------------------------------------
*Mondeca**
*3, cité Nollez 75018 Paris France
Web:    www.mondeca.com <http://www.mondeca.com>
----------------------------------------------------
Tel:       +33 (0) 871 488 459
Mail:     bernard.vatant@mondeca.com <mailto:bernard.vatant@mondeca.com>
Blog:    Leçons de Choses <http://mondeca.wordpress.com/>

Received on Thursday, 24 January 2008 15:09:27 UTC