Caveat: I'm not familiar with TGN...
I would presume that if the value "Kinder Scout" is a defined
taxon within the taxonomy (hierarchy) of geographic names according
to the TGN type, that such hierarchical information would be defined
elsewhere, probably in a standard, definitive URN namespace scheme
and the official RDF schema for TGN (or a particular version of TGN).
And, following the argued (at least by me ;-) ideal that known,
standardized enumerations be defined as URN identified resources,
it would be preferable to have something like
<rdf:Description
rdf:about="http://chris.croome.net/photos/2000/Walk_up_Kinder_Scout_16_Janua
ry/29_Saifi_and_Bill_on_the_top_of_Kinder.jpg">
<dc:coverage rdf:resource="urn:tgn:Kinder+Scout"/>
</rdf:Description>
and elsewhere (such as in a standardized schema)
<rdf:Description rdf:about="urn:tgn:kinder_scout">
<rdf:label>Kinder Scout</rdf:label>
<rdfs:seeAlso
rdf:resource="http://vocab.pub.getty.edu/cgi-bin/tgn_browser/tgn.spl?key=110
5226&searchtype=record&file=index.html"/>
<!-- additional definitions regarding hierarchical positioning, etc. -->
</rdf:Description>
In fact, it would be expected that the URN namespace scheme for TGN
would explicitly reflect the hierarchical structure of the taxonomy, e.g.
urn:tgn:world/north_america/us/fl/orange/kinder_scout
(just using bogus and certainly wrong values -- haven't a clue where Kinder
Scout is, certainly not in Orange County, Florida ;-) -- but you get the
point...)
Then, if there also happened to be a "Kinder Scout" elsewhere in the world,
there is a clear distinction in the URNs as to which is meant, etc. even if
they have the same label. And of course, one can also define labels/names
according to other criteria such as language, historical period, etc. simply
by qualifying the rdf:label nodes.
As mentioned above, it is likely that versioning would have to be reflected
in the URN namespace scheme, to ensure persistent validity of references,
e.g.
urn:tgn:v1/world/north_america/us/fl/orange/kinder_scout
urn:tgn:v2/earth/na/us/fl/orng/kinder_scout
And use RDF to define equivalence relationships between members of different
versions which represent the same "thing" but have different URNs.
Eh?
The key is simply to get standards bodies responsible for controlled
vocabularies
and enumerations to define (or allow others to define) official, definitive
URN schemes and values which allow those managed abstract resources to be
referenced in metadata efficiently, and most importantly in a consistent
fashion by "everyone", and allow knowledge about members of those controlled
vocabularies/enumerations to be defined once, not redundantly throughout the
semantic web for gazillions of property value nodes.
Cheers,
Patrick
--
Patrick Stickler Phone: +358 3 356 0209
Senior Research Scientist Mobile: +358 50 483 9453
Nokia Research Center Fax: +358 7180 35409
Visiokatu 1, 33720 Tampere, Finland Email: [log in to unmask]
> -----Original Message-----
> From: ext Chris Croome [mailto:[log in to unmask]]
> Sent: 27 September, 2001 16:42
> To: [log in to unmask]
> Subject: Re: Poor man's structured value construction versus explicit
> reso urceURI values
>
>
> Hi
>
> On Wed 26-Sep-2001 at 06:16:52PM +0300, Patrick Stickler wrote:
> >
> > since lots of folks in a hurry just copy and modify examples (myself
> > included ;-) perhaps some alternate resource based examples
> and clear
> > verbage would be a good thing.
>
> I agree that some more examples would be really helpful :-)
>
> One thing I'm not sure how to do is a qualified Spatial
> Coverage using the
> Getty Thesaurus of Geographic Names. For example the RDF for
> a photo taken on a
> hill near me -- would it be something like this?
>
> <rdf:Description
> rdf:about="http://chris.croome.net/photos/2000/Walk_up_Kinder_
> Scout_16_January/29_Saifi_and_Bill_on_the_top_of_Kinder.jpg">
>
> <dc:coverage>
> <dcq:TGN>
> <rdf:value>Kinder Scout</rdf:value>
> <rdfs:seeAlso rdf:resource
> ="http://vocab.pub.getty.edu/cgi-bin/tgn_browser/tgn.spl?key=1
> 105226&searchtype=record&file=index.html" />
> </dcq:TGN>
> </dc:coverage>
>
> How would it's Hierarchical Position be represented?
>
> Chris
>
>
> --
> Chris Croome
> http://mkdoc.com/
> http://www.webarchitects.co.uk/
>
|