Poderopedia: Who is who in business and politics in Chile
Name of the organisation: Poderopedia.org
Category: Data journalism website or section small media
Description of the project: Poderopedia is an editorial and crowdsourced data journalism website that uses public data, it`s own relational intelligence methodology, semantic web technology, it`s own ontology and network visualizations to map who's who in business and politics in Chile. Its aim is to promote greater transparency and to become a tool that helps journalists do better journalism, with richer data, more context, and visualizations that make massive amounts of connections and information accessible to any citizen. So far Poderopedia has created profiles of 2310
persons, 880 companies and 588 institutions and we keep creating new ones everyday. The motivation behind this comes from the belief that because information is power, everyone should be entitled to that information in order to help people make more informed decisions; shed light on eventual conflicts of interest; make powerful people, companies and institutions more accountable of their acts; and promote better democracy and transparency. The project is non profit, open source and intends to provide it`s platform for newsrooms and journalist so they can create their own mapping and research projects.
What data did you use and how did you obtain it?: In countries like Chile, what`s your last name and where you are born define much of your chances in life, because the elite are very much inbred. All this determines who becomes powerful and influential in the business and political world, usually affecting (for better or worse) the course of public affairs, government decisions, and bills approved in Congress.
Journalism has always been about asking the right questions at the right moment with the best contextual information in order to dig deeper and discover new things that are relevant to the public. But with an overflow of information, it's hard to distinguish between what's urgent and important. Sometimes media companies don't have the time or money for it. Many don`t care. In other cases, editors are just worried about tomorrow and think of developers as the guys who fix their computers.
In this mega real-time data ecosystem, there is a big opportunity for curating, sharing, hacking and linking, as writers Clay Shirky and Jeff Jarvis and others have said so many times. It's a chance to provide filtered content, meta-content and context-driven content to help people understand what happened, why, and who is responsible for it.
Everything has to do with who's connected with whom and how, whether it's a company merger that leaves thousands out of work, the approval of a hydroelectric project rejected by the community, or the decision to pass a bill that will neglect our children's education but will serve corporate interests.
All this data is out there. It's in articles written by journalists, in documents, wikis, and public databases provided by government offices, in studies made by scholars or NGO reports, in information given by whistle-blowers that reaches the hands of citizens. These are the sources Poderopedia feeds on.
Poderopedia feeds from public validable sources such as media, open data from chilean government and Congress sources, public databases, crowdsourcing and it`s own reporting. Our editorial team feeds the platform and validates every information that is sent by users and collaborators.
Technical challenges, tools used: Creation of our own connections methodology, creation of our own ontology and semantic technologies to store and query data. These technologies allow us to represent a diverse set of relations between entities (people, companies, organizations) in a flexible way. The goal is to identify and express relations of power and influence of people and organizations. The connections and ontology allows us to mark down all entity relations and turn them into automatic data visualizations using D3.js. We also use Documentcloud to store Documents and REST technology on top of Python and Web2Py and RDF.