[CITE-Forum] Axis order issue in WFS 1.1 cite tests

Andrea Aime aaime at openplans.org
Mon Sep 3 11:29:31 EDT 2007

Andrea Aime ha scritto:
> Hi,
> the cite test data has the following two interpretations of the
> same feature:
> Postgis version:
> INSERT INTO "PrimitiveGeoFeature" VALUES ('description-f001', 
> 'name-f001', NULL, geometryfromtext('POINT( 39.73245 2.00342 )',4326), 
> NULL, 155, 'http://www.opengeospatial.org/', 12765, NULL, '2006-10-25Z', 
> 5.03, 'f001');
> GML3 version:
> <sf:featureMember>
>     <sf:PrimitiveGeoFeature gml:id="f001">
>       <gml:description>description-f001</gml:description>
>       <gml:name 
> codeSpace="http://cite.opengeospatial.org/gmlsf">name-f001</gml:name>
>       <sf:pointProperty>
>         <gml:Point gml:id="g003" 
> srsName="urn:x-ogc:def:crs:EPSG:6.11.2:4326">
>           <gml:description>description-g003</gml:description>
>           <gml:pos>39.73245 2.00342</gml:pos>
>         </gml:Point>
>       </sf:pointProperty>
>       <sf:intProperty>155</sf:intProperty>
>       <sf:uriProperty>http://www.opengeospatial.org/</sf:uriProperty>
>       <sf:measurand>1.2765E4</sf:measurand>
>       <sf:dateProperty>2006-10-25Z</sf:dateProperty>
>       <sf:decimalProperty>5.03</sf:decimalProperty>
>     </sf:PrimitiveGeoFeature>
>   </sf:featureMember>
> Now, I just had a discussion with some OGC referecing experts, namely
> Carl Reed and Arliss Whiteside, about how to interpret refencing
> systems. Both say that whilst EPSG:4326 is uncertain due to the WMS 1.1
> saying it's lon/lat and the WFS 1.3 spec saying the opposite,
> the URN form has been created explicitly to avoid this confusion
> and it's unmistakeably lat/lon. 

So, nobody answered this one... yet, what I reported is a solid bug
in the WFS CITE tests sample data sets. Either that, or what
Carl Reed and Arliss Whiteside told me does not match with what the WFS 
expert group is thinking.

What is the position of the WFS expert group and cite test managers?
Is "urn:x-ogc:def:crs:EPSG:6.11.2:4326" to be considered a lon/lat
CRS? Or is it the sample data broken? And if so, which version of
the coordinates is the right one?


More information about the CITE-Forum mailing list