Re: Standard HTTP Header for Geolocation

Fernando: I think you might be in the wrong place.  This group doesn't do RFCs.  You may want to look here:
<http://tools.ietf.org/wg/geopriv/>
The IETF GEOPRIV working group did look at working on an HTTP header some time ago, but never actually finalized anything.  As you note, though, that was several years ago, and now might be a good time to raise the issue again.  You're welcome to submit an Internet-draft.

Of course, the W3C also does work defining HTTP headers, so maybe it could be added to the scope of this WG as well.

It does seem like a Geolocation HTTP header could be a useful optimization.  It's already the case that:
1. Several current UAs allow persistent permissions for a site, and 
2. Many sites send Geolocation back in XHRs already

Given these conditions, it seems like you could save an HTTP round-trip to the client by simply having the UA send a Geolocation header with requests to authorized sites, rather than the page having to send it back in an XHR.  If you wanted to be slightly less liberal, you could also define an option in the API that would allow pages to request this functionality.  

--Richard





On Feb 20, 2011, at 12:34 PM, Fernando Ribeiro wrote:

> Hello,
> 
> Has this working already evaluated submitting a RFC for a standard HTTP header for geolocation information, as suggested in its charter?
> 
> There was a draft some years ago (http://tools.ietf.org/html/draft-daviel-http-geo-header-05), when handful things like the geo: scheme (http://geouri.org/) hadn't been developed yet.
> 
> With location-based applications being more common by the hour, I think this header is urgent to minimize the number of incompatible implementations in the wild.
> 
> Thanks,
> 
> Fernando

Received on Monday, 28 February 2011 19:15:49 UTC