W3C home > Mailing lists > Public > public-xg-geo@w3.org > January 2007

Re: [foaf-dev] FOAF, geonames, and more

From: Benjamin Nowack <bnowack@appmosphere.com>
Date: Fri, 26 Jan 2007 17:56:26 +0100
To: "Alexandre Passant" <alex@passant.org>
Cc: public-xg-geo@w3.org, foaf-dev@lists.foaf-project.org
Message-ID: <PM-GA.20070126175626.CFB16.1.1D@>

Hi Alexandre,

RDFS is "descripive, not prescriptive", i.e. if you describe
a thing as being based near some other thing, then you simply
assert that they are spatial things. There is no problem with
using based_near with an org if that org is a spatial thing.


On 26.01.2007 17:07:06, Alexandre Passant wrote:
>Hi all,
>I'm currently looking for ways to link things (especially foaf:Agents)
>to geonames [1] defined entities (since it provides not only lat/long
>info but also useful things as wikipedia entry, neighbour places ...).
>Currently, there's a foaf:based_near property, which links two
>Since geo:Feature (which is used to define all geonames.org ontology
>instances) and foaf:Person are subclasses of SpatialThing, we can link
>any foaf:Person to a geonames.org "Feature" as:
>http://apassant.net/foaf.rdf#me foaf:based_near
>Yet, foaf:Agent itself is not a subclass of geo84:SpatialThing
>So, I cannot say that a foaf:Organisation is located in a city.
>One solution could be to subclass not only foaf:Person from
>geo84:SpatialThing, but foaf:Agent itself.
>The other idea would be to create a new property as locatedIn (in
>geonames or in geo84 ontology, or even new namespace) that will link
>any owl:Thing / rdf:Resource (and not only subclasses of SpatialThing)
>to a geo84:SpatialThing, and that then can be subclassed:
>(could also create the property in an existing namespace, and the
>subproperties in another one, as the relationship vocabulary [2])
>Thus, I'll be able to describe the location (and specify which kind,
>i.e. workplace, establishement place ...) of any foaf:Agent, but also
>other things as a ical:VEvent, to point to the city the event is
>located, using any geonames Feature instance.
>(I could have suggest to subclass foaf:based_near, but as the "near"
>notion is relative, I'm not sure that's a good idea to create
>subproperties from it)
>I'll be happy to get feedback about these ideas.
>foaf-dev mailing list
Received on Friday, 26 January 2007 16:57:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:43:25 UTC