[Geojson] [OSGeo-Standards] EPSG + Coordinate Ordering (Was Re: GeoJSON '1.0'?)

Christopher Schmidt crschmidt at metacarta.com
Fri Mar 14 04:30:31 PDT 2008

On Thu, Mar 13, 2008 at 07:41:17PM -0400, Frank Warmerdam wrote:
> Christopher Schmidt wrote:
> >That was not the plan for GeoJSON. The additional flag would not change
> >the data in any way, all it would do would be to accept, and describe,
> >the mapping from EPSG:4326 to the way GeoJSON works. The spec was (is)
> >very clear that the coordinate order is *always* x, y, z.
> Chris,
> No offence, but "x, y, z" is not as specifically meaningful as you might
> think.  Do you mean "easting, northing, elevation"?  There are real 
> coordinate
> systems that routinely encoding x as southing, and y and westing
> (for instance).

And the CRS definition explicitly defines what x, y, and z are in the GML
definition for the projection. This was brought up by mpd: he offered
as an example, which uses x, y as southing, westing:


However, I have not yet found a Cartesian CS definition which does not
define the mapping from the directions in the data to x, y: regardless
of what it means, data must pick *some* coordinate system, and in order
to understand it in the world, you will have to map it to a CRS. The CRS
explains what X, Y, Z means.

Is there some coordinate reference system for which this isn't true?
Everything I've seen defines a <CoordinateSystemAxis> <axisAbbrev /></>
which provides the association, but I'll admit I only picked out a few
esoteric projections and went based on those, and have way less
knowledge about the subject than you and many others do.

Christopher Schmidt

More information about the GeoJSON mailing list