Open Tech Projects

The elections will be held in five phases on April 16, April 22, April 23, April 30, May 7 and May 13, and the results will be announced on May 16. The timelines here and on all ther to-do pages have been based on the election dates. 



UI :

  1. On some pages like here the text is bleeding into the sidebar.
  2. On individual blog posts, the sidebar doesn't show up.
  3. Change the colour of the dots for all incidents to a colour other than RED. Decrease the size of the dots so that the entire map doesn't look full.


  1. Fix the outgoing alert functionality/ page. When you select a city, or a location, you should be given a custom RSS feed, which you should be able to subscribe to, by RSS or email. The same RSS feeds would be pointed to  the citywise Twitter/ SMSGupShup accounts (like VoteMumbai). The present UI is fine, except that the feed URL doesn't change. If this is not possible, remove the map and point to our single feed and Twitter A/c. - WIP (Munish / Gautam)
  2. Integrate e-mail reports into the singular backend admin interface. - WIP (Munish / Gautam)
  3. Simulate Swift integration on by embedding RSS feeds into the Ushahidi interface. In the UI we will replace the default "mainstream media" section and expose streaming links, microbloggery, videos, and photos about the election, using our growing sources list. References are, and the wireframe here. Someone needs to parse the incoming feeds, then put them in a table and display them with a WHERE clause in sql (as per andrew). - WIP (Prajwala)
  4. Localize in Hindi.
  5. Fix location 'Find' in backend admin interface when using non-Google maps.



  1. Decide on the "validation" model (how many people does it involve, and what is the UI)
  2. Remove the Ushahidi "approval" process (replaced with the new validation model)
  3. Privately experiment the new sweeper interface and validation model output
  4. Public launch sweeper UI ready for volunteers
  5. BUG: Error validation in the submit report page prevents from saving a report. Example: Enter  "||R5BgSxs=" in the "News source link" input field.




  1. Location extraction and tagging on all incoming urls People invovled:  UI: Chris / Backend: Andrew




  1.   - Analytics and NLP for automated clustering
  2.   - Full integration and structured publishing of people and places
  3.   - more advanced sweeper UI with entity recognition and auto-suggest tagging.
  4.   - user testing.
  5.   - ability to evaluate importance (eg, retweet analysis)
  6.   - Test implement Dawid Weiss' code for clustering links (Carrot:


People Involved: UI and user testing: Chris / Backend: Andrew / Analytics Support: Sean Gourley / developers / David Weiss code (perhaps implemented by Anselm Hook)


 BY MAY 13





For swift-specific development please see