[georss] Intended lat/long order

[Ron Lake]

Hi,

 

Not a problem. My inattentiveness caused most of it.  The issue of
lat-long order is a great source of debates, because so many people seem
to think there is a "correct" or a "natural" one.  There is not.  There
is only the one specified by the CRS you claim you are using - no more
and no less.  So if you CLAIM you are using what EPSG (now OGP) calls
4326 then the order MUST be (lat,lon).  You could just as correctly use
(long,lat) and many people/programs do - but then you cannot claim this
is EPSG 4326.  Coordinate order is determined by the coordinate
reference system (more accurately the coordinate system part of it).

 

Cheers

Ron

 

From: georss-bounces at lists.eogeo.org
[mailto:georss-bounces at lists.eogeo.org] On Behalf Of
rico.hauke at daimler.com
Sent: November 9, 2007 12:18 AM
To: georss at lists.eogeo.org
Subject: Re: [georss] Intended lat/long order

 


Hi, 

Sorry for all the trouble/confusion I've caused. I actually always
thought that lat-long is the correct order *until* I came across this
post on postgis-users which totally confused me: 
(First three lines after the quotes.) 

http://postgis.refractions.net/pipermail/postgis-users/2007-October/0174
53.html 

I am providing GeoRSS feeds and store them in a PostgreSQL/PostGIS
database. So now I know for sure, that the order in GeoRSS is lat long,
I just have to figure out if the PostGIS WKT Reader really expects long
lat? 

Thanks, 
Rico 






creed at opengeospatial.org 

11/08/2007 06:39 PM 

To

rico.hauke at daimler.com 

cc

jlieberman at traversetechnologies.com, rsingh at opengeospatial.org 

Subject

Re: [georss] Intended lat/long order

 

		





Rico - 
  
Your first example. 
  
The axis order in a GML encoding is defined by the Coordinate Reference
System (CRS). The OGC (and just about every vendor and standards
organization) now uses the EPSG database as the normative reference for
all CRS parameters. GeoRSS GML defaults to WGS 84 2d as defined in the
EPSG database. As such, the order is lat-long. 
  
Believe it or not, axis order is one of the most contenscious topics in
the geospatial and standards communities. However, the OGC, OASIS, IETF,
ISO, and others when referencing WGS 84 all use lat-long as this is the
axis order defined by the CRS entry in the EPSG database. Most GIS
systems have historically not followed this convention. Hence the
controversy.  FYI, the next version of SQL-Server implements OGC's
Simple Features SQL - and the axis order is lat-long and they are using
WKTs. Finally, FYI, KML 3.0 will follow the correct EPSG convention and
switch axis order from long-lat to lat-long. This is not common
knowledge, so please do not spread this item of information around yet.
Thanks 
  
Hope this helps. I suspect others will weigh in. 
  
Regards 
  
Carl Reed, PhD 
CTO 
OGC 
----- Original Message ----- 
From: rico.hauke at daimler.com 
To: georss at lists.eogeo.org 
Sent: Thursday, November 08, 2007 6:51 PM 
Subject: [georss] Intended lat/long order 


Hi list, 

This might sound stupid, but I was wondering what the lat/long order
(eg. in a point geometry) is intended to be? 

So if I am located at latitude 37.4 and longitude -122.1 which of the
following points are valid GeoRSS representations? 

1. lat/long order: 
<gml:Point>
      <gml:pos>37.4 -122.1</gml:pos>
</gml:Point> 

2. long/lat order: 
<gml:Point>
      <gml:pos>-122.1 37.4</gml:pos>
</gml:Point> 

The reason I am asking is that I have to pass the position string into a
WKT StringReader which wants the input to be in x/y (=long/lat) format. 

Thanks, 
Rico 

________________________________

_______________________________________________
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/20071109/8f9b97a8/attachment-0001.html