+1 from me either.

A further note: this applies to all the POWDER-S properties used in the 
attribution (i.e., wdrs:validFrom, wdrs:validUntil, dcterms:issued). 
Another consequence is that we cannot use dcterms:issued, since in [1] 
it is defined as a subproperty of both dc:date and dcterms:date, and 
also its range is defined (RDFS Literal).

This means that we have to define a new property: wdrs:issued


Phil Archer wrote:
> On yesterday's telecon [1], it was resolved that
> "That the range of wdrs:issuedby should be defined such that its range 
> can be foaf:Agent or dcterms:Agent but can also be other (unspecified) 
> classes"
> It was further agreed that we would stick with treating POWDER-S 
> documents as OWL Ontologies (which are disjoint from OWL Things). A 
> POWDER-S document therefore typically begins like this:
> <owl:Ontology rdf:about="">
>   <wdrs:issuedby
>     rdf:resource="" />
>   <dcterms:issued>2007-12-14T00:00:00</dcterms:issued>
> </owl:Ontology>
> Stasinos made the point that anything that describes an OWL ontology 
> (and therefore anything that describes a POWDER-S document) must be an 
> OWL Annotation property and not anything else. foaf:maker, for example, 
> has a domain of owl:Thing and therefore _cannot_ (legally) be used to 
> say who created a POWDER-S doc.
> Following yesterday's meeting, Andrea spotted more text in the OWL 
> documentation that Stasinos confirmed was entirely relevant [2]. The 
> bottom line is that it seems the way forward for POWDER is that we do 
> not define a range for wdrs:issuedby. WE _do_ however, want to encourage 
> people specifically to use FOAF or DC Terms hence my proposal that the 
> line in section 2.1 of the DR doc that defines issuedby [3] be changed to:
> "Exactly one issuedby element MUST be included and it MUST use the src 
> attribute to point to an RDF resource that describes the entity that 
> created the POWDER document. It is  RECOMMENDED that this be done using 
> an instance of the Agent class from either the FOAF or Dublin Core 
> vocabularies."
> Further, that we do not define a range for wdrs:issued
> Stas, Andrea - have I reflected the position accurately?
> Everyone - do you agree with the proposal?
> Phil.
> [1] (member only link)
> [2] 
> (member only link)
> [3]

Andrea Perego
Dipartimento di Informatica e Comunicazione
Universita` degli Studi dell'Insubria
Via Mazzini, 5 - 21100 Varese, Italy

Received on Tuesday, 7 October 2008 15:55:07 UTC