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

Hi all,


Jo Walsh wrote:
> dear Alexandre,
> On Fri, Jan 26, 2007 at 05:07:06PM +0100, Alexandre Passant wrote:
> 
>>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 ...).
> 
> 
> This sounds like an interesting and fun project. I would be wary of
> over-relying on geonames.org - the US mil/gov/int data from which is it
> derived can be very inaccurate in places and have names which are locally
> meaningless sometimes. Though geonames are refining and fixing it slowly...
> 
> 
>>I'll be happy to get feedback about these ideas.

FWIW, I think linking an agent to a location is not enough. You proposed 
e.g. also linking events to geonames entities, I think the in fact 
alltogether it makes most sense:

geospace - event - agent

for instance in foaf, based_near is (probably on purpose) defined very 
vaguely. but even there, I would like to tie to "events",

e.g. I lived from my birth until the age of three in vienna,
.... I lived from 2003 til jan 2006 in innsbruck and, I live in madrid 
since jan 2006.

all this is based_near information, but I can't express all of it in
foaf.

actually, a temporal component would be a great extension to be tackled.

Maybe to theoretical of nature, but temporal RDF [1], might be a cool 
option, though linking via ical events is fine eas well.... btw: how do 
I model ongoing in ical?

best,
axel

1. http://www.dcc.uchile.cl/~cgutierr/papers/temporalRDF.pdf

p.s.: this thread is a bit cross-posting (besides foaf-dev and 
public-xg-geo, I also included expertfinder-dev now, shame on me... 
suggest that dicussion continues on foaf-dev, but probably also 
interesting for the others)

> I can offer you pointers to prior art as i'm a bit short of screentime
> 
> http://esw.w3.org/topic/GeoInfo has a ton of biscuit trails to it on
> vocabs/use cases. there was a 'pim' personal info management ns that defined 
> a 'near' property and also a lot of use of the 'cyc:near' property.
> i guess foaf:based_near sort of superseded that. There are so many
> subtleties once you start to get into the spatial ontology side ("does
> a point really exist? is time relative? OMG, what am i sitting on?")
> that being usefully vague is about the most you can hope for :)
> 
> I wrote a "mapping FOAF networks" piece in "Mapping Hacks" which
> probably says most of the same things, it might be worth you looking
> at if you run across a copy, or i might be able to dig out the
> original word file. Jim Ley did a lot of stuff using 'nearest airport'
> as a predicate and you can get a surprisingly long way to
> interestingness with that. I move so much, i need a bot to remind me
> where i am.
> 
> best luck,
> 
> 
> jo
> _______________________________________________
> foaf-dev mailing list
> foaf-dev@lists.foaf-project.org
> http://lists.foaf-project.org/mailman/listinfo/foaf-dev
> 
> 


-- 
Dr. Axel Polleres
email: axel@polleres.net  url: http://www.polleres.net/

Received on Monday, 29 January 2007 18:03:46 UTC