Re: First release of geocode-glib available

[Bastien Nocera <hadess hadess net>]

On Tue, 2011-05-17 at 09:21 +0300, Alberto Mardegan wrote:
> On 05/16/2011 01:42 PM, Bastien Nocera wrote:
> > On Sun, 2011-05-15 at 12:44 +0300, Alberto Mardegan wrote:
> >> On 05/14/2011 11:24 PM, Bastien Nocera wrote:
> >>> The fact that we have only one implementation means that we have one
> >>> well maintained implementation. I don't think it's much of a problem. Do
> >>> you have particular reasons why you think that supporting multiple
> >>> backends is actually useful?
> >>
> >> Availability of up-to-date street data is heavily dependent on the provider.
> >
> > Your point being?
> 
> My point being that supporting multiple backends is definitely useful: OSM, 
> Yahoo, Google, Nokia maps and others can have very different degrees of 
> coverage. Some areas which are shown as streetless by Google have quite detailed 
> street information in OSM, and vice-versa.
> Also, even if two map providers have detailed street informations for one area, 
> it may be that one is more up-to-date than the other.

That also means more services to follow, and more code to maintain.
Right now, it seems to work fine for most of our uses, and even if
there's no street-level geocoding for quite a few countries, there's
still higher level geocoding:
http://developer.yahoo.com/geo/placefinder/guide/requests.html#supported-countries

Which is probably good enough for showing a map in Empathy, a Twitter
client, or fetching the nearest airport code, or even the place's
timezone.