Friedrich Lindenberg

pudo.org

Social network analysis for advocates and journalists.

On June 21st, the Knight News Challenge Round on Data ended. The day before, Rufus, Ross and I sat down to write out some ideas that we'd been discussing for a while. The first idea I want to repost here is a proposal for Grano, which I've discussed in this blog before.

1. What do you propose to do? [20 words]

We’ll make a powerful tool for journalists and advocates to keep track of actors and their relationships in complex environments.

Grano

2. How will your project make data more useful? [50 words]

It’ll enable users to manage research in a structured way, helping them to link raw data to the actors, events and organisations they’re already investigating and to find those that they may have missed before. We’ll help users do their job more thoroughly, while creating a structure that can be re-used later.

3. How is your project different from what already exists? [30 words]

Network analysis means many things to people: it’s graph algorithms to coders, network diagrams to designers and CRM to business. Journalists and advocates need evidence gathering and information linkage to be at the core of these things.

4. Why will it work? [100 words]

We want to focus on four functions that will make this a practical tool instead of a gimmick:

a) allowing users to easily integrate bulk data to complement manually entered information,

b) helping them to keep track of the source for each fact that is entered and keeping a full version history,

c) providing easy access control so that users can choose which information to keep private and which links to publish with others and

d) text snippets, so that researchers can combine structured analysis and narrative fragments in which the tool will detect references to the network’s entities.

5. Who is working on it? [100 words]

The Open Knowledge Foundation wants to cooperate with investigative networks around the world to develop this project. We’ve already been pioneering data collection and presentation tools, such as DataHub.io and OpenSpending, as well as efforts like the Data Journalism Handbook and the School of Data to widen data literacy.

  • Friedrich Lindenberg (OKFN) has worked on several data projects and data-journalism training and will lead this project.
  • Ross Jones (OKFN) will contribute as a software architect.
  • Stefan Candea (2011 Nieman Fellow at Harvard, Director of the Romanian Center for Investigative Journalism) has offered to advise us.

6. What part of the project have you already built? [100 words]

We’ve already built Grano, a REST backend that can store network information, generate custom reports about nodes and relations and run full-text search. Because we think that meaningful network analysis is hard, we are conservative in the choice of technology to focus on outcomes. To force that, we decided to base our tool on a concrete use case. The software is now first used in an unannounced project that tracks lobbying in the EU, powering a special-purpose, JavaScript-only site. Unfortunately, this means the current prototype does not have a stand-alone web interface and the serious data integration capabilities we think it needs.

7. How would you use News Challenge funds? [50 words]

We want to develop Grano to give investigative journalists and civic hackers a (re-usable) web interface to design their network structure, manually enter data, integrate bulk data sets and to explore the resulting network, make notes, calculate key metrics and to export reports, rankings and network visualizations.

8. How would you sustain the project after the funding expires? [50 words]

While the service is going to be of immediate use, we believe that advocacy groups and newsrooms will also deploy it as a backend to their features and campaign sites. We aim to make Grano into a thriving open source project, supported through custom services for power users.

If you like this idea, please vote for it on the proposal page.