[GeoJSON] Content-type

[Erik Wilde]

hello sean.

thanks for the feedback.

On 2013-11-05, 08:52 , Sean Gillies wrote:
> It seems like the well established pattern on the web is
> application/xxx+json: see GitHub's application/vnd.github.v3+json et al.
> Flipping the hierarchy around could add confusion.

it's not just a pattern, it's an RFC. a new one and so far not a widely 
used one, but still, it's a standard.

> Looking at http://tools.ietf.org/html/draft-bray-i-json-00#section-5,
> the new I-JSON now *does* have room for profiles. So the future of being
> more precise about GeoJSON could look like "application/json;
> profile=http://geojson.org/profiles/collection", where
> http://geojson.org/profiles/collection identifies a profile that extends
> I-JSON and has type: FeatureCollection at the top.

http://www.ietf.org/mail-archive/web/apps-discuss/current/msg10708.html 
might be interesting to follow. personally, i think:

- GeoJSON should have a well-defined media type of its own, following 
the structured suffix spec (application/geojson+json);

- this media type should support the profile media type parameter; and

- profiles could the be used to identify certain specializations of 
GeoJSON, following the original idea of what profiles are for as 
described in http://dret.typepad.com/dretblog/2013/03/on-profiles.html

cheers,

dret.

-- 
erik wilde | mailto:dret at berkeley.edu  -  tel:+1-510-2061079 |
            | UC Berkeley  -  School of Information (ISchool) |
            | http://dret.net/netdret http://twitter.com/dret |