3 ms·

Hi - it's actually quite different. As of now, we don't (yet) have a forward geocoding api exposed. Rather, most of the API functionality that's live is for bac

by othman 17y ago

Hi - it's actually quite different. As of now, we don't (yet) have a forward geocoding api exposed. Rather, most of the API functionality that's live is for backward geocoding. As in, with Google's Geocoder, you give a string and you get back a place. With us, you give a place (i.e. lat/lon) and you get where that place is and what's around it. That being said, a number of people have been pinging us about getting simple geocoding functionality, so stay tuned for that as well :).

othman 17y ago
BTW, there are also a number of other differences - below is a quick list: - Type of data: One of things we do is combine a number of different datasets (16+ M businesses, points of interest, intersections, etc...) - Query type: You get very powerful expressiveness in the types of queries you do. You can literally make a request for 'Japanese restaurants that are within 1 mile and that are open at 10pm.' - Media layers: For most entities, you can get media layers such as Twitter and Flickr. For example, you can get tweets around the mission (http://api.geoapi.com/v1/e/mission-san-francisco-ca/view/twitter?apikey=demo&pretty=1 http://api.geoapi.com/v1/e/mission-san-francisco-ca/view/twi...). We're adding more layers as we go. - UserView: Each developer gets a private namespace into which they can add information and run geo queries on it. Basically, you can annotate the world with close to zero effort.
- lilypad 17y ago
  The query syntax is pretty cool, but as far as raw geocoding I don't really get it -- Google has a much bigger dataset than this, offers reverse geocoding just fine (With good neighborhood data in the US and Europe). And I gather you are just proxying requests to twitter and flickr (and not actually storing all this for all places yourself) -- why wouldn't I just query their APIs directly? Not trying to be snarky, I just don't understand....
bbsabelli 17y ago
Huh? New York: http://maps.google.com/maps/geo?q=40.724039,-74.00176 http://maps.google.com/maps/geo?q=40.724039,-74.00176 What am I missing here?

by othman 17y ago

othman 17y ago

BTW, there are also a number of other differences - below is a quick list: - Type of data: One of things we do is combine a number of different datasets (16+ M businesses, points of interest, intersections, etc...) - Query type: You get very powerful expressiveness in the types of queries you do. You can literally make a request for 'Japanese restaurants that are within 1 mile and that are open at 10pm.' - Media layers: For most entities, you can get media layers such as Twitter and Flickr. For example, you can get tweets around the mission (http://api.geoapi.com/v1/e/mission-san-francisco-ca/view/twitter?apikey=demo&pretty=1 http://api.geoapi.com/v1/e/mission-san-francisco-ca/view/twi...). We're adding more layers as we go. - UserView: Each developer gets a private namespace into which they can add information and run geo queries on it. Basically, you can annotate the world with close to zero effort.

lilypad 17y ago
The query syntax is pretty cool, but as far as raw geocoding I don't really get it -- Google has a much bigger dataset than this, offers reverse geocoding just fine (With good neighborhood data in the US and Europe). And I gather you are just proxying requests to twitter and flickr (and not actually storing all this for all places yourself) -- why wouldn't I just query their APIs directly? Not trying to be snarky, I just don't understand....

bbsabelli 17y ago

Huh? New York: http://maps.google.com/maps/geo?q=40.724039,-74.00176 http://maps.google.com/maps/geo?q=40.724039,-74.00176 What am I missing here?