[rdfweb-dev] Re: Getty Thesaurus of Geographic Names

[Danny Ayers]

>Ian Davis and I used the Getty Thesaurus at Calaba and we have a fair
>bit of experience of working with it's idiosyncrasies. We were
>disambiguating place names in hierarchical web sites, looking for
>e.g. London in South Africa, and classifying pages accordingly.

Sounds great.
Heh, I half expected it was known, but I didn't expect that anyone here
would actually have worked with it...


>TGN is available in a machine readable format, from which can be
>generated. But the license is expensive (was $10K for commercial
>usage) and the licensing terms disallow the republishing of the
>thesaurus in a verbatim form - i.e. it's okay to use it for internal
>disambiguation, but not okay to publish in RDF through a web
>interface like xmlns/wordnet.

Ouch. That is prohibitive, shame.

>It's pretty complete (250K place names worldwide?) but for one
>country [1] we know about, annoying in it's accuracy, which can be
>counter-intuitive.

I'd also consider Leeds just as a city in (East?) Yorkshire, though I
vaguely remember it being catalogued somewhere as the 'Leeds-Bradford
Conurbation', like a Greater London/Manchester entity. As you say,
counter-intuitive.