[georss] GeoRSS Location References

[Raj Singh]

Even better. I only kept <georss:where> around the GML link because I  
remember you having problems conforming to the abstract model and/or  
creating RDF without it. But if that's not an issue, it's certainly  
better to get rid of it.

Let me anticipate your next comment. You might say that <georss:where>  
belongs in the referenced document, and now that I'm not being lazy  
and actually taking a look at the schema (http://georss.org/xml/1.0/georss.xsd 
), I agree. In that case, maybe I should change the MIME type from the  
standard GML one to "application/georss-gml+xml"?

---
Raj


On Apr 2, 2008, at 5:59 PM, Joshua Lieberman wrote:
> Raj,
>
> A link to a GML simple feature (or any other feature) makes sense as
> an alternative or authoritative reference, but putting the link inside
> georss:where seems a bad idea for both definition and processing. A
> feed processor can no longer assume that there is a very simple
> geometry inside the element to run with, plus it is ambiguous what the
> difference is between links inside and outside. It would be a mess in
> RDF, where it would be a property in one context and an object in the
> other. I think it would be best to define usage of a link in the
> normal place in the entry / item and restrict georss:where contents to
> their present georss geometry (profile in draft).
>
> --Josh
>
>
> On Apr 2, 2008, at 5:29 PM, Raj Singh wrote:
>
>> Just posted a location reference proposal that I hope crystallizes  
>> the
>> consensus we've been moving to. It covers GML and Simple, and handles
>> referencing an XML child element. It's silent on the multiple
>> geometries issue, but I think it will support the eventual solution
>> nicely (or conversely, any multiple geometries solution should have  
>> no
>> trouble playing nice with this).
>>
>> Referencing External Geometry with Atom link
>> http://georss.org/node/31
>>
>> ---
>> Raj
>>
>> _______________________________________________
>> georss mailing list
>> georss at lists.eogeo.org
>> http://lists.eogeo.org/mailman/listinfo/georss
>
> _______________________________________________
> georss mailing list
> georss at lists.eogeo.org
> http://lists.eogeo.org/mailman/listinfo/georss