Re: [georss] GeoRSS meeting review

Dan,

Thanks for your questions. Some responses below.

Josh

On Apr 2, 2007, at 12:44 PM, Dan Brickley wrote:

> Joshua Lieberman wrote:
>> Dan,
>> Please give us some feedback on http://www.w3.org/2005/Incubator/ 
>> geo/Wiki/Geo_Update so that we can move ahead with this RDF/OWL  
>> vocabulary, which matches GeoRSS. There are many aspects of "how"  
>> to realize a vocabulary in these languages which are not entirely  
>> straightforward to optimize, so input is really needed.
>
> I'll try to take a proper look. First question ... it mentions ISO  
> 19109,
>
> "NeoGeo (proposed) consists of a root property _featureproperty  
> which takes as its domain any OWL/RDF class that it makes sense  
> (after ISO 19109) to cast as a geographic feature."
>
> Is ISO 19109 freely available? If not, could someone summarise? The  
> W3C "basic geo" vocab defined a class "spatial thing", which seems  
> to be in a similar role. But we didn't go into much detail about  
> what we considered that to exactly amount to. Could the domain of  
> "featureproperty" be basicgeo:SpatialThing, for example?

ISO 19109 the document has a fee like other ISO documents (I will  
never understand this...). However, many of the ISO 191nn specs have  
their equivalents in the OGC Abstract Specifications ( http:// 
www.opengeospatial.org/standards/as ) and in this case, the model is  
described in Topic 5: Features.

We have discussed further the "roots" of NeoGeo. Although a geometry  
should not itself be a feature, one possibility will be that both can  
be considered SpatialThing's. In that case, the domain of  
_featureproperty is _Feature, but the range of georss:where is  
_Geometry. Whether _Feature is necessarily spatial is another  
question, but this will do for now.

               Thing
                  |
        SpatialThing
        /                   \
_Feature         _Geometry
           \               /
        georss:where

This is the latest version I have locally.

>
> I can't currently get to http://mapbureau.com/neogeo/neogeo.owl and  
> the host doesn't seem to respond to pings.

This is a question to Chris Goad (of course, it has been a reliable  
link up until now). Chris, could you perhaps include the ontology  
itself on the wiki, or should I do that?

>
> cheers,
>
> Dan
>
> Dan