[georss] GeoRSS Location References

[Ron Lake]

-----Original Message-----
From: Christopher Schmidt [mailto:crschmidt at crschmidt.net] 
Sent: March 30, 2008 9:59 PM
To: Ron Lake
Cc: Raj Singh; GeoRss
Subject: Re: [georss] GeoRSS Location References

 

On Sun, Mar 30, 2008 at 09:39:03PM -0700, Ron Lake wrote:

> Hi,

> 

> In GML the type of the resource referenced in the link is signaled by

> the remoteSchema attribute.  This is usually an xpointer expression
that

> references a schema fragment describing the resource.  It is up to the

> application to decide if it wants to fetch the remote schema fragment

> and validate the remote resource.

 

I'm not entirely sure if this is a similar situation. It sounds somewhat

dissimilar. In any case, it doesn't 

 

It doesn't what?  I understood that you wanted to know the geometry or
other type of the resource referenced in the link.

 

> In RDF all such references (rdf:resource) must comply with the
resource

> type identified in the schema (RDFS).  RDF parsers always fetch and

> validate such links.  

 

Huh? Neither of these are true in RDF. 

 1. RDF is 'open world': an RDFS can describe only one particular

    interprtation, and the basic assumption is always that there are

    other interpretations out there. (Possibly even conflicting ones.)

 

I confess that I have not used RDF parsers for a while, but those that
enforce a RDFS have to fetch the remote resource - and ones I used did
do this and checked the type.

 

 2. No RDF parser I've ever used has ever fetched anything I didn't

    explicitly tell it to.   

 

> This was felt to be too heavy weight in GML.

 

Regards,

-- 

Christopher Schmidt

Web Developer

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.eogeo.org/pipermail/georss/attachments/20080330/a7aaaebd/attachment.html