[GeoJSON] spec clarification of CRS

Raj Singh raj at rajsingh.org
Fri Dec 13 07:50:44 PST 2013


On Dec 12, at 5:01 PM, Martin Daly <Martin.Daly at cadcorp.com> wrote:

>> That's all fine and totally clear if you stop there. Or even if you go on to say,
>> "and by the way there's an implied CRS of urn:ogc:def:crs:OGC:1.3:CRS84"
>> there. I only have the problem with adding in a CRS property that is allowed
>> to specify an axis order of y, x. Maybe just take one question at a time:
> 
> We are not hard of understanding, so feel free to ask as many questions as you like in a single email.
> 
>> 1) does EPSG:4326 specify y, x axis ordering?
>> Does anyone have a definitive answer to this?
> 
> Having said that, I fail to see where this particular question is going.
> 
> I do not recall ever seeing anything on this list, or IRC, or wherever, that disputes that some CRS-s - including  all EPSG-defined geographic CRS-s - have axis order that is *not* easting/longitude followed by northing/latitude.
> 
> Once again, all that GeoJSON says is that the encoding of coordinates is *always* easting/longitude followed by northing/latitude. GeoJSON does not say that CRS-s that do not use this order are disallowed, only that encoding coordinates in other orders is disallowed.
> 
> Interpreting the enforced coordinate order as meaning that some CRS-s are therefore disallowed appears to be your, and OGC-s, interpretation; but that interpretation is simply wrong.
> 

This is just me speaking -- not as OGC staff, as no OGC group or member asked me to investigate this question. OK I finally understand what's going on. Now I get that I shouldn't interpret the enforced coordinate order as meaning that some CRS-s are disallowed. I disagree with that decision -- it doesn't make any sense to me to allow the use of part of a CRS definition -- but at least now I understand what's going on. 

---
Raj



More information about the GeoJSON mailing list