This application is a story-writing tool for reporters who want to be provided with context while composing their drafts. Based on a draft text, it will find the name of companies and people, retrieve open data based on these names, and show you previous notes you collected on any of them.

This is an implementation of text-snippet based SNA; the idea that data in a social network isn't necessarily well-structured but can also be little pieces of text that link to each other (and to entities).

Original mockup: here.

This app is an experiment in whether such a semi-structured approach to influence and story mapping can be used to capture the building blocks of journalistic investigations.

Before you can install storyweb, the following dependencies are required:

• A SQL database. While we recommend Postgres, the app can also run with other databases, such as SQLite.
• ElasticSearch for full-text indexing.
• less, installed via npm.
• Python, and Python virtualenv.
• RabbitMQ

Once these dependencies are satisfied, run the following command to install the application:

git clone https://github.com/granoproject/storyweb.git storyweb
cd storyweb
virtualenv env
source env/bin/activate
pip install -r requirements.txt
python setup.py develop
npm install -g bower uglify-js

Next, you need to customize the configuration file. Copy the template configuration file, settings.py.tmpl to a new file, e.g. settings.py in the project root and set the required settings. Then export the environment variable TMI_SETTINGS to point at this file:

cp settings.py.tmpl settings.py
export STORYWEB_SETTINGS=`pwd`/settings.py

Use bower to install javascript dependencies:

bower install

To create a new database, run the following command:

python storyweb/manage.py initdb

This will also create an admin user with the email address [email protected] and the password admin which you can use to log in and create more users.

Congratulations, you've installed storyweb. You can run the application using:

python storyweb/manage.py runserver

This tool is heavily inspired by Newsclip.se, a hack from the Al Jazeera "Media in Context" hackathon in December 2014. Thanks to the team: Eva Constantaras, Kasia Dybek, Bruno Faviero, Heinze Havinga, Friedrich Lindenberg, Phillip Smith.

It is licensed under an open source MIT license. We welcome any contributions to the code base.