[Geojson] coordinate order

Christopher Schmidt crschmidt at metacarta.com
Sun Mar 9 07:31:26 PDT 2008

On Sat, Mar 08, 2008 at 10:00:27PM -0800, Dale Lutz wrote:
> Hi,
> For the record, I too think this is a great change.  I really like the way you've kept the "x is x" and still allowed those that really want swapped coordinate systems to eat their cake too -- they just need to be explicit about it.

Note that this doesn't change the data inside GeoJSON at all: it's only
additional metadata about correlating CRS order with coordinate order. 

> FME indeed does support GeoJSON, and in fact at our user conference yesterday we were showing FME Server spitting GeoJSON into OpenLayers (Peter mentions this in his blog here
> http://geothought.blogspot.com/2008/03/fme-user-conference-report.html).  FME 2008 is finalized now so it won't be able to correctly read any data that has the x/y flipped, but we'll get that into the next FME betas very shortly, and we don't write flipped x/y with FME 2008 so whatever we produce will still work fine as well.

Again, so long as you're just working in x/y, the coordinate order in
the data won't change: the only thing that will change is that you now
have a wider range of options for how you describe the CRS. Before, you
were limited in what you could do: now you're not. I don't know how much
this affects FME: I don't know the level of support you put into using
CRS-based coordinate ordering. I'm hopeful that this is just a case
where if you understand coordinate ordering from CRSes, you simply read
the CRS information from your dictionary and write out a correct
coordinate_order member into your CRS object.

> I'll ask around at Safe to see if there's anything that we'd add to the current draft proposal, but our guys had already gone through things carefully and I fully expect that we'd heartily support calling the proposal a 1.0 as well.

Christopher Schmidt

More information about the GeoJSON mailing list