[rdfweb-dev] locations?

Dougal Campbell dougal at g...
Mon Jan 6 14:53:24 UTC 2003


On Mon, 6 Jan 2003, Jim Ley wrote:

> "Dougal Campbell" <dougal at g...>
> > I finally tracked down a useful reference:
> > http://dublincore.org/documents/2000/07/28/dcmi-point/
> >
> > This definitely looks like a better starting point. Any recommendations
> > for adding the other attributes (postal code, state/province,
> > hometown/workplace/summer-home, etc)?
> >
> > How about:
> >
> > <point:Point name="Enterprise, AL">
> > <point:north>31.3</point:north>
> > <point:east>-85.85</point:east>
> > <rdf:Description>
> > <dc:description>Hometown</dc:description>
> > <contact:City>Enterprise</contact:City>
> > <contact:StateOrProvince>AL</contact:StateOrProvince>
> > <contact:Country>us</contact:Country>
> > <contact:PostalCode>36330</contact:PostalCode>
> > </rdf:Description>
> > </point:Point>
>
> Since you don't mention the projection, and Dublin Core's default is to
> wooly to be used, and only lets you define it as a string literal anyway, I
> don't like this for lat/lon. As it's much easier if we can get people to
> agree on a projection we should by preference use a predicate which limits
> users to one, they can still then use others.

The default units and projection are signed decimal degrees on the
globe.

> Also, contact:stateOrProvince looks dodgy, using a region specific 2 letter
> code without specifying the region. (although maybe since country is also
> defined more unambigously we're okay)

Right, I figured that in combination with "country", "stateOrProvince"
should be okay. But I'm definitely open to alternatives. The important
thing is that it should be acceptable to the general FOAF population and
also standard enough for machine digestion and usability.

-- 
Ernest MacDougal Campbell III, MCP+I, MCSE <dougal at g...>
http://dougal.gunters.org/ http://spam.gunters.org/
Web Design & Development: http://www.mentalcollective.com/
This message is guaranteed to be 100% eror frea!




More information about the foaf-dev mailing list