[Geojson] Three more things...

[Christopher Schmidt]

On Sat, Mar 15, 2008 at 07:38:01AM +0000, Martin Daly wrote:
> 1) Add a *requirement* that features must have either exactly the same  
> CRS as their parent featureCollection, or none at all. the latter  
> implying using the featureCollection CRS (where present)

I woul go even farther: "CRS Must be defined on the uppermost parent of any
GeoJSON object": so if I get something, and it has a type: "Feature",
"FeatureCollection", or anything else thate GeoJSON defines... that's
where my CRS is defined. Thoughts? 

> 2) Add a *recommendation* that date properties are encoded by  
> producers as ISO8601 strings, and parsed back into dates by clients  
> (where possible)

ISO 8601 strings look like:
1994-11-05T13:15:30Z

For those people among us who aren't bright enough to know that off the
top of my head.

The "parsed back into dates by clients" is something we've avoided: we
don't make any processing recommendations in the spec for encoding. I'm
not strongly for or against the date property encoding recommendation.

> 3) Add a *recommendation* that all features in a featureCollection  
> share the same set of properties (different values of course), to fit  
> into the near-universal relational model

Hm, I don't like this personally: it gets too close to the 'predefined
schema' stuff that I've tried to stay away from, and makes some 
FeatureServer datasources and OSM data 'against' recommendation. I feel
like this is not required as part of the spec. But that might just be
because I'm one of the weirdos who's going to do it anyway :)

Regards,
-- 
Christopher Schmidt
MetaCarta