[GeoJSON] Unique media type for GeoJSON?

[Erik Wilde]

hello sean.

On 2014-05-07, 11:55 , Sean Gillies wrote:
> application/json+geojson would go against the grain of media types that
> are already out there:

yes, http://tools.ietf.org/html/rfc6839 says that it should be the other 
way around...

>      application/vnd.github+json (GitHub)
>      application/vnd.api+json (JSON API)
>      application/vnd.collection+json

...and all of these do it the right way.

> We could register application/vnd.geo+json with IANA (following
> instructions at http://tools.ietf.org/html/rfc6838) which looks
> straightforward. We could add application/geo+json to the geojson I-D
> (http://datatracker.ietf.org/doc/draft-butler-geojson/) but then we're
> blocked until that makes it to RFC. I think I prefer the former.
> Any advice on the above, Erik?

the latter looks a bit cleaner, but in the end there's no real 
difference: it's an opaque identifier using a '+json' suffix. so i think 
looking at timing issues might be a good way to decide.

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 |