Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Crowdsourced situational awareness toolkit.
JavaScript Ruby Java
Pull request Compare This branch is 20 commits ahead, 1 commit behind ajturner:master.
Fetching latest commit...
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
spec Added rspec.


Welcome to Swift River aka "Swift" aka "crowdsourced goodness" aka "that fork of VoteReport"
based on the VoteReport '08 app

Official Docs Page:

Other super-relevant links you probably should visit:

    * The Swift Mailing List: "":
    * The Swift Homepage: "":
    * The Swift repository: "":

An overview of the Open Source Software project known as Swift:

Swift builds on Twitter Vote Report, a platform for crowd-sourced election monitoring. Twitter Vote Report made it possible for US voters to report voting irregularities in real-time using a standard set of tags for monitoring wait time at polling places. Swift hopes to expand this approach into a general purpose toolkit for crowdsouring the semantic structuring of data so that it can be reused in other applications and visualizations. The developers of Swift are particularly interested in crisis reporting ( and international media criticism (

The community around Swift seeks to develop software for "Citizen Editors" and support civic digital activity with crowdsourced interaction designs for an Open-Data-based discussions, research and visualizations.

Our design philosophy is to develop a completely nonpartisan and white label application using Open Source tools. Our work is intentionally minimalist, intended to be included in larger nonprofit efforts such as and We hope to provide a complete, standalone, unbranded application that is capable of aggregating and re-syndicating articles. We are using Linked Data tools such as to enrich our archives with entity tags, which can be human corrected and potentially extended with situationally-relevant tag vocabularies.

How do we hope to do this? 

The Swift interface serves a very narrow and well-defined role in the ecosystem of crisis software: the "triage" of data that emerges during a crisis. Swift engages self-interested teams of “citizen editors” who curate publicly available information about a crisis or any event or region as it happens. We hope that Swift will be reusable in many important contexts, particularly in realtime during a crises and disaster. We hope to find a meaningful and managable way to volunteer "in your pajamas" during an event, by participating as a volunteer "Sweeper," curating and correcting data as it emerges about the crisis or event.

Coming Soon: A guide to creating and managing a Swift Project. (Currently you have to just install your own copy of the application. This will be fixed soon as per our roadmap in the docs.)

Excited yet? Read your eyes out:

    * Documentation: Hosted on Github with the source code:
    * Vote Report application code: This is the core of existing app. Swift is a fork of Vote Report, currently with very minor additions. We’re hacking it into something even better, and foraging a lot of other great FOSS software along the way.
    * Carrot third party code: A clustering engine and UI. So you can put in data and get back thematic categories. Useful for lots of unstructured data. see:…
    * Solr/Lucene third party code: Handle the search backend.
    * Freebase, Geocommons, Wikipedia and OpenStreetMap APIs will serve as our structured repository of data whenever possible using the glue in the taxonomy editor project.
    * Github Source Code Managment: we are doing all of our work through Github. You can do an initial checkout by installing git and running: git clone git://

I want to setup an instance and help hack!

    Please contact either organization if you are interested in being a development partner. For support please chime in on the discussion list at 

But I Really Want More Technical Concepts and Scope:

Ok, Here's a small but complete statement of what the Swift project intends to do:

    * Users can create Instances, such as “Ushahidi: Swine Flu Tracking” or “Meedan MENA News” which are RSS aggregators
    * Instances bundle RSS feeds from numerous sources, such as
    * When users create an instance they become administrators for that instance, and can add or delete sources.
    * An instance is then available for sweepers to join and being tagging links as they emerge from the sources.
    * The tags used can be arbitrary or can be part of a more complex taxonomy of tags, which is editable by the administrators of an instance.
    * The veracity of links is computed.
    * The resulting set of scored and annotated links is available via API

The Swift Stack:

    * Strong reliance on the existing link-sharing and taggable ecosystem: Twitter, Flickr, Delicious and YouTube integration.
    * Rails with RESTful API
    * MySQL
    * Lucene/SOLR.
    * Complete Test Suite
    * Strong Semantic Web integration (using taxonomies via on
    * Git-based verion control
    * Amazon EC2 LAMP hosting

Implemented modules:

  Polls for Tweets

  Polls YouTube API for relevant keywords and grabs links to videos.
  Polls Flickr API and grabs links to videos.

  Email text, photos, or audio to ""

    Mozes for SMS messages that contain configurable keywords like:
      #dctrip, #inaug09, #dctrip09, #inaug

    Accepts phone calls into an Asterisk system.
    See adhearsion/components/vote_report/lib/votereport.rb for the geeky details

    Accepts audio, photo, and text reports from the Inauguration Report dedicated iPhone application

    Accepts reports from the Inauguration Report dedicated Android application



Made available under the MIT License.

Copyright 2008-2009 by the following authors and contributors, based on the VoteReport '08 app:

David Troy,
Andrew Turner,
Billy Gray
Cory Forsyth
John Trupiano
Bryan Liles
Brendan O'Connor
Deanna Zandt
Nat Freitas
Chris Strom
Tony Herrera
Chris Blow
Something went wrong with that request. Please try again.