A Human-Centered Framework for Redesigning Public Debate Online

Vibe investigates how we might make discussions online better and more sustainable through the use of personalized data and pro les that right now are only used for the purpose of advertising. We’re speculating on how, without collecting any additional data, but just reusing the one that was already collected more interesting debates might emerge. Looking at how data can be used to enable the rich expression of thoughts, how it might introduce people to the discussion that have not participated before, and how we might use computational techniques to combat abuse and harassment in a respectful way.





Designing for Meaningful Contribution 


People want to contribute to a discussion by putting great content forward. This can mean everything from telling a great story to sharing knowledge or contributing personal data to improve an article.

  • Interactive Sparkline is a simple yet universal and exible micro-interac- tion that enables this type of rich sharing of content of any kind.
  • A simple tool for the Journalist allows them to target specific audiences they want to hear from using the same technology that traditional ad targeting tools are based on. The same piece of software also allows the journalist to explore data contributed by readers.

We believe that this concept can have signi cant impact because it lets users share stories and experiences, while it reuses data that is already there in a clever way. Additionally it potentially motivates users to give the system more data which bene ts the social network. 


Designing for Diversity



The filter bubble or the echo-chamber are among the most discussed problems in journalism of the last years. Algorithms designed to only show us things that we might like further seperate use from each other instead of connecting us. What if we could use the same targeting-data to show the whole discussion, bring together seperate discussions of the same topic and even motivate people to join the discussion who are under-represented. 

  • Conversation starters analyze the unique perspective you could have on a topic based on previous experiences and nudge you to join the conversation. 
  • A simple summary-view enables you to discover interesting topics and get the gist of the discussion 

We believe that this concept can have significant impact because people have a desire to leave their echo-chamber and have real, open discussions. This will make the platform more attractive, which means more time spent on the platform which is a significant metric for social networks. 


Designing for Augmented Moderation 



Moderation in online discussion is often very opaque without any ac- countability. While the system works to some extent, people who put a lot of effort into contributing to the discussion too often nd themselves having their comment deleted due to some liberally applied community guideline by an automated system or a human curator.

What if we could enable a more nuanced review process that fundamentally relies on automation but can hand off to a human auditor if necessary.


  • Quarantine serves two important purposes. On the one hand it is an easy way to manage the ranking of potentially harmful posts without outright deleting them. On the other hand it serves as some sort of user support forum to trou- bleshoot a potentially mis-reported posting directly inline. It offers an easy user interface to complement both automated bots but also hand off to human auditors. 

We believe that this concept can have signi cant impact because people want to feel safe to express their opinions online without the fear of being harassed or attacked. This will make platforms more attractive. At the same time the reduction of human auditing will save costs and make the system more scalable in the long run. 

You can check out our prototype for this concept on Twitter,