[georss] elevation and Simple<->GeoRSS GML interop questions

[Joshua Lieberman]

Chris,

The gmlgeorss12.xsd file is for a profile of GML 3.1.1, where the  
coordinate dimensions are (more) closely  tied to the coordinate  
reference system, so one needs generally to declare both a 3D  
coordinate system (compound or not) and dimensions defined by that  
system.

The schema allows a georss:where and georss:elev to be used together  
(yes, they are properties at the same level). I think we've discussed  
discouraging it, but not sure it is actually banned outright. It's a  
conceivable use case, I suppose, of needing specific 2D information  
and only general elevation information, but awkward at best.

The two schemas georss.xsd and gmlgeorss12.xsd is distinct. The latter  
is a proper profile of GML 3.1.1, meaning that any XML which validates  
under it will also validate under the full GML 3.1.1 schema. You are  
free to use those elements just as you are free to use GML. The former  
schema defines the elements which are specific to GeoRSS, which import  
and use GML elements and are "not" themselves part of GML.

GeoRSS as a specification only has something to say about what one  
communicates with XML which uses GML within the GeoRSS elements, so if  
you just use elements alone from gmlgeorss12.xsd, you are on your own  
in defining what your application schema or XML elements mean relative  
to other information.

There has been an interest expressed in creating GML application  
schemas (having a feature element derived from gml:AbstractFeature)  
where the geometry properties are constrained to those used in GeoRSS  
(In other words, valid according to gmlgeorss12.xsd). A paper  
describing that will hopefully be released by OGC in June or so.

Cheers,

Josh

On May 6, 2008, at 4:45 PM, Chris Beaudette wrote:

>
> 1. I don't see an elevation spec for GeoRSS GML in gmlgeorssXX.xsd,  
> but there's one for Simple in georssXX.xsd (<georss:elev>).  Looking  
> through the user forums at http://lists.eogeo.org/pipermail/georss/2007-January/000936.html 
>  I see that it is necessary to use the "dimension" attribute for the  
> <gml:pos[List]> tag, e.g.
>
> <gml:LineString>
>    <gml:posList dimension="3">0.0 0.0 1000.0 ...</gml:posList> </ 
> gml:LineString>
>
> Where the '1000.0' is the elevation.
>
> - Is this correct for gmlgeorss12.xsd?  I don't see the 'dimension'  
> attribute in there. I see only 'srsDimension'.
> - If this is the correct use of 'dimension', what units are they  
> in?  Does it depend on the 'srsName' used?  Does 'dimension' only  
> work when using a 'compound CRS'? Or are a compound CRS and  
> 'dimension' not used together?
> - Would we ever still have occasion to use both GeoRSS GML and  
> Simple together, like this (fromhttp://lists.eogeo.org/pipermail/georss/2008-April/001848.html) 
> ...
>   <georss:where>
>     <gml:point>
>       <gml:pos>blah blah</gml:pos>
>     </gml:point>
>   </georss:where>
>   <georss:elev>313</georss:elev>
>
> ... assuming the <georss:elev> tag is at the same level as the  
> <georss:where> tag, or this if it is not:
>   <georss:where>
>     <gml:point>
>       <gml:pos>blah blah</gml:pos>
>     </gml:point>
>     <georss:elev>313</georss:elev>
>   </georss:where>
>
>
> 2. For an application to use the XML components only of the GeoRSS  
> GML spec w/o Atom (and the requisite <entry> tag), would it be  
> possible to use the <gml:...> elements without the <georss:where>  
> tag?  Or is <georss:where> required even when not using Atom/<entry>?
>
>
> 3. As a follow-up to #s 1 & 2, I see the whereType defined in  
> georssXX.xsd, but not in gmlgeorssXX.xsd. Is it necessary to use  
> both schemas when using GeoRSS GML structures only (and no Simple  
> features)?  Or is that only necessary when using the <georss:elev>  
> tag w/ GeoRSS GML?
>
> _____________________
> Chris Beaudette
> Windsor Solutions, Inc.
> www.windsorsolutions.com
> 503.675.7833  ext. 232
>
> _______________________________________________
> georss mailing list
> georss at lists.eogeo.org
> http://lists.eogeo.org/mailman/listinfo/georss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.eogeo.org/pipermail/georss/attachments/20080506/19c518dd/attachment.html