Please sign in to comment.
A little bit of documentation...
... everyone writes documentation when it's actually self promotion ;)
- Loading branch information...
Showing with 42 additions and 0 deletions.
|@@ -0,0 +1,40 @@|
|+Where in the World?|
|+This project is a quick experiment to see whether a maps-style interface can improve accessiblity of|
|+some of the Guardian's travel writing. Some of the content is already geo-tagged, so this app just|
|+takes those geotags and links to the articles from a google map. It scans for new content in the API|
|+and will include that automatically, though Google heavily cache the map overlays so I'm not sure what|
|+the latency is yet.|
|+There are around [33,000 items of guardian content tagged travel](http://content.guardianapis.com/travel/travel),|
|+though many of these are not geotagged.|
|+The API only lets you query for 50 at once, and cannot (currently) filter on only geotagged items.|
|+So, a background process scans for modified content, and maintains a database in MongoDB. The runtime app|
|+uses this database to build a [KML file](http://code.google.com/apis/kml/documentation/)|
|+The app itself is implemented in Scala, using [akka](http://akka.io) for background scheduling,|
|+and [salat](https://github.com/novus/salat) to talk to mongo, [the lift web framework](http://liftweb.net) and the Guardian's|
|+[Content API scala client](https://github.com/guardian/open-platform-content-api-scala-client).|
|+Right now, it's hosted - for free - by [CloudBees](http://www.cloudbees.com) and [MongoHQ](http://mongohq.com).|
|+You can see it running at [here](http://travel-map.tackers.cloudbees.net/).|
|+Although I work for the Guardian, and helped implement the Open Platform Content API, this|
|+is entirely a personal project and has no connection at all with the Guardian.|