>
> Harry,
>
> I mean something like:
>
> <rdf:Property rdf:about="http://purl.org/dc/terms/audience">
> <rdfs:label xml:lang="jp">[Label in Japanese]</rdfs:label>
> <rdfs:comment xml:lang="jp">[Definition in Japanese]</rdfs:comment>
> <xyz:translates rdf:resource="http://dublincore.org/usage/terms/history/#audience-001"/>
>
> ..where xyz: is either a DCMI-maintained vocabulary or
> something like OWL.
Dear Tom,
the problem i have with the "non-reification" approach is,
that one asserts the PROPERTY "audience" translates "audience-001" -
when one in fact has in mind, that one translates labels and/or
comments.
Suppose you collect 15 different translations of 9 differnt versions of
term-declarations.
How would you know, which translation is a translation of version 4?
The translation property in my view should go to the
statements "...and a label in Japanese is" and
"... a comment in Japanese is"
something like:
<owl:Ontology rdf:about="">
<owl:imports rdf:resource="http://purl.org/dc/terms/"/>
<owl:imports rdf:resource="http://purl.org/dc/elements/1.1/"/>
</owl:Ontology>
<rdf:Property rdf:about="http://purl.org/dc/terms/audience">
<rdfs:label rdf:ID="audi001label" xml:base="http://my.org/#" xml:lang="de">Zielgruppe</rdfs:label>
</rdf:Property>
<rdf:Description rdf:about="http://my.org/#audi001label">
<dct:references rdf:resource="http://dublincore.org/usage/terms/history/#audience-001"/>
<dc:creator rdf:nodeID="B"/>
<dct:issued>2003-11-17</dct:issued>
</rdf:Description>
<abc:Org rdf:nodeID="B">
<rdfs:seeAlso rdf:resource="http://www.my.org"/>
<rdfs:label xml:lang="de">Meine Organisation</rdfs:label>
<abc:address>NoWhereLand, NoWhereTown, NoWhereLane 51</abc:address>
</abc:Org>
>
> In other words, the declaration of the translation of a term
> would carry an extra assertion pointing to the specific
> historical version in English used as the basis of the
> translation. To be clear: I do _not_ mean the translation
> should be "about" the specific historical version instead of
> the generically identified entity "Audience".
>
> Now, if the Japanese site _also_ were to give a version
> identifier to each successive revision of the _Japanese
> translation_, then one might want to add yet _another_
> assertion, such as:
>
> <dcterms:hasVersion rdf:resource="http://www.ulis.ac.jp/dublincore/#audience-jp-001"
>
> following the model used for the declaration of Audience (in English)
> at http://dublincore.org/2003/03/24/dcq#.
>
> Tom
>
> On Fri, Nov 14, 2003 at 10:41:51AM -0500, Harry Wagner wrote:
> > > In that case, I would like to suggest the model of identifying
> > > particular historical versions of a term with a separate URI, as
> > > in:
> > > http://purl.org/dc/terms/audience
> > >
> > > the "term URI" supported by the DCMI Namespace Policy, as opposed to:
> > >
> > >
> > http://dublincore.org/usage/terms/history/#audience-002
> > http://dublincore.org/usage/terms/history/#audience-001
> >
> > the successive historical states of that term. In other words,
> > a Japanese translation of the term "Audience" could include the
> > assertion:
> >
> > Translates: http://dublincore.org/usage/terms/history/#audience-001
> >
> > Tom,
> > I'm not sure I follow you. Are you recommending a practice different from
> > what we do ourselves? Looking at this from the perspective of one of the
> > distributed sites, this might be difficult to swallow.
> >
> > The Japanese example you give uses a dublincore.org URL. Is this correct?
> > Would not this be URL that resolvs to a server in Japan?
> >
> > Regards... hw
>
> --
> Dr. Thomas Baker [log in to unmask]
> Institutszentrum Schloss Birlinghoven mobile +49-160-9664-2129
> Fraunhofer-Gesellschaft work +49-30-8109-9027
> 53754 Sankt Augustin, Germany fax +49-2241-144-2352
> Personal email: [log in to unmask]
>
>
|