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

Frank Warmerdam warmerdam at pobox.com
Thu Mar 13 07:23:34 PDT 2008

Christopher Schmidt wrote:
> Members of the OSGeo standards list:
> In discussion of the creation of the GeoJSON specification, we're
> running into a question of whether you can refer to something as
> EPSG:4326 and 'override' the coordinate order, insisting that it is
> x,y[,z] regardless of what EPSG says. Apparently this came up before in
> WMS 1.3.0 discussions: if anyone has more context, please respond to
> this message on the osgeo-standards list  (I will summarize the contents
> back to the GeoJSON list).


The GeoJSON specification can override anything it wants from the EPSG
definitions.  The axis ordering changes with WMS and other OGC specifications
were not exactly because ISO said it had to be.  It was more (IMHO) that
different people with different priorities in the ISO process were able
to force a change.

I will note that the EPSG database distribution license does not permit
changes of axis order in the database itself.  But that does not preclude
applications from only using some of the information (for instance ignoring
axis order info), nor does it preclude offering overrides which is the route
I've taken in GDAL and PROJ for stuff I want to alter from EPSG on an
individual coordinate system basis.

I will say there are some modest pitfalls to ignoring axis order from the
EPSG database.  In particular there are projected coordinate systems where
the axis order is important, and where the axis order in practical use
in the world is not "easting, northing".  So your concern is just the
lat/long vs. long/lat issue with geographic coordinate system then perhaps
you should be very specific about that.  For instance state that the axis
order is overridden to be "long, lat" for all *geograhic* coordinate
systems but that for projected or other coordinate systems EPSG axis
order will be honoured.

If you are trying to specify coordinate systems with something like
"EPSG:4326" you might even consider having an altenate form to
to indicate your standard overridden form.  Something like
"EPSG_LONGLAT:4326".  Of course, the danger is that some folks
will be sloppy and still use EPSG:4326 and apps will accept it and
it will become practice anyways.

I will note it is slightly rude to EPSG to use their geographic
coordinate system definitions and ignore their axis order definitions.
At least something like "EPSG_LONGLAT:4326" makes it more clear that
you aren't using their definition directly.

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org

More information about the GeoJSON mailing list