Version Tomorrow is the first day of the rest of your life
lecture: The quest for algorithm diversity
Our vision is to increase transparency behind personalization algorithms. facebook.tracking.exposed, is two things: a web browser add-on that collects the stories visible in your Facebook Newsfeed, and a website that aggregates that data from all the users who have the addon installed.
Our mission is to help researchers and users assessing how current filtering mechanisms work.
Personalization algorithms should be modified in order to minimize the dangerous social effects of which they are indirectly responsible and to maximize the values, both individual and social, that algorithms should incorporate.
facebook.tracking.exposed is a tool developed before Donald Trump made the filter bubble a mainstream problem. Filter bubble is one of the side effects of personalisation algorithm (P.A.) and P.A. have a certain impact in our society, we can assume they are nothing less than social policies. Bad side: social policies have to be debated openly and not in a technical meeting intended to optimize revenues.
Increasing recently we listen to theories on how facebook and google can be “regulated”, but the civil society has not yet a scientifically reliable tool to judge what PA are actually doing.
The project has many innovative and techno-political aspects:
- see what is happening in the social media, with the eyes of another user
- try to understand the logic of an algorithm, assess what is prioritize and why. try to guess if a content is discriminated positively or negatively
- offer open data about our metadata collected, show how companies can embrace data transparency and privacy-aware publications of public data
The project is currently in ß phase, at SHA2017 will be released a 1.0 version