Bubble Burst - An App To Make Online Discussions A Little More Enlightened


Why are public discussions on the internet often so terribly void of meaning and full of foul language?

Everyone can discuss anything at all times online today. Think about how the Trump administration has been discussed, how international affairs are being discussed. Anything really. Yet, much of those discussions on social media or even the comment sections on reputable news outlets are highly polarized and divisive. This is sad because everyone being able to discuss anything, anytime, should really be a great precondition to find the will of the people and improve our democracies. Why is it not?

Firstly, in any public discussion, people like to look good. This regularly leads to displaying high status or dominating contributions involving shaming the other and/or alleviating oneself, provoking a response in kind leading down to a dead end.

Secondly, our mind is wired to love coherence in concepts, memory and self. It is more pleasant and requires less effort to process information consistent without our existing beliefs. To top it all off, our minds also love stories more than facts - even when we falsely generalize from those stories. As a result, urban myths, non-fact based arguments or even fake news with a great story easily pollute public discussions.

Thirdly, if you look at a discussion forum or the comments after a news article or (facebook) post, there is usually no clear order, multiple threads run in parallel and reading it is cumbersome.

Fourthly and finally, at the end of the discussion no one takes note of what the audience has actually written and the policy itself is again made by those politicians that follow their own agenda. So why bother - as an apathetic voter might argue?

In sum, our egos leading to dominating or even aggressive posts, our cognitive biases, the lack of organization and the perceived lack of influence of those discussions make them unpleasant and unhelpful for our democracies.

Welcome screen (1).jpg
Discuss - Select two needs.jpg

What are the 4 ingredients for the "perfect democratic discussion" then?

First off, to position yourself in the discussion, a non-violent statement of your interest/need would be much better than shaming/attacking others, e.g., instead of "I hate immigrants" something like "I need to feel safe in my city despite a lot of people coming in that I do not know." Or instead of "We must leave the NATO which is an evil war machine" something like "I want us and our government to be non-violent despite a world in which war is a reality".

This accomplishes two things:
1) Not shaming/attacking another to not provoke a response in kind
2) Framing your interest/need explicitly and independently from only one issue which allows more room in the discussion to come up with creative “win-win” solutions that accommodate multiple interests, e.g., “showing solidarity with refugees while taking care of (perceived) safety needs.”

Secondly, assuming that the interests / needs are on the table and the discussion has been phrased as such. e.g., "How can we feel safe and show solidarity regarding refugees?" or "How can we feel safe and be non-violent in a world in which war is a reality?" (note how these questions are much broader than what is usually discussed) the evaluation of facts and logic can proceed (over fake news or overgeneralizations from stories).

Thirdly, the discussion needs to be organized well by threads that follow a particular argument, evidence or policy issue. A simple tagging and thread system should solve that. Additionally, as users may be interested to read both pro and contra contributions to receive a balanced feed, it makes sense to develop the feature of tagging by the contributor as a solution. Finally, upvoting of contributions by users and tracking what contributions users read the most should make for powerful organization (some of the elements mentioned already work effectively in forums such as stackoverflow or forums where people submit change requests for software products).

Fourthly and finally, it would make sense for participants of the public discussion to be able to submit requests based on the additional insight they have generated, for example to request parties/politicians to take a stance on an issue or media or research institutes to evaluate a fact or likely impact of a policy.

Discuss - Select a discussion.jpg

What and how to prototype this app/platform idea?

The grand project would be to build one global platform for democratic discussions incorporating all four ingredients. However, this would likely fail because of the huge effort of curating content and building a user base. Instead, a lean approach would attempt to iteratively gain as much learning as possible with the least possible effort.

Safely left out can be the element allowing you to make a request to media or politicians because essentially anyone can already do that today (too little learning). More learning but a huge effort and, hence, a bad learning/effort ratio would be the fact and logic checker. This is because fact checking a single contribution in an automated way is technically still very challenging (high effort) and single attempts at rating the trustworthiness of news outlets and/or single contributions already exist (medium learning).

We are left with "not shaming/attacking another and framing your interest/need explicitly” and “organizing discussions well along with filters”. Both appear to provide plenty of learning but the organization of the discussion would require less effort (in coding) than convincing users to state their needs and not attack others. At the same time, not attacking others may be a necessary precondition for a discussion not to be hijacked by haters or angry rants. So let’s assume for a moment that both elements need to be piloted jointly (a final thought on this later).

To pilot both elements, most importantly, a discussion with sufficiently engaged users is needed. Instead of starting that discussion from scratch, we could go where discussions already take place. Three options would be 1) the comment sections on news websites such as NY Times, BBC, Al Jazeera etc., 2) discussions around referenda or local politics, e.g., virtual extensions of town hall meetings 3) discussions set up by initiatives such as democracy labs, Respekt im Netz (Respect on the internet) or liquid democracy that specifically want to build better and more democratic discussions using technology.

Of the three options, the initiatives attempting to make better discourse possible are likely the most accessible and most receptive to our approach. Also, they are more likely to attract open minded individuals willing to state their interest/need explicitly refraining from shaming/attacking. This would reduce the effort further. Hence, they would constitute a good pilot. If successful, one could continue with local politics or referenda which are more contentious and more effort would be required to find people willing to start with a vulnerable, non-violent statement first and keep others from shaming/attacking. At the same time, one would also create more value for democracy in this context.

Hand - Rate the discussion.jpg
Hand - Create and post request.jpg

Come on board if you agree or tell us why you disagree

So, if you agree that more explicit statements of one’s own interest/need, instead of attacking/shaming others and a more open-minded search for solutions, well organized by threads and filters to find useful contributions on both sides, will help us build better public discussions, come join us. Seriously. We need a team. If you disagree with the analysis of the problem or the proposed solution, tell us why. I am sure it would teach us a lot.

If you’re curious and would like to find out more, feel free to drop me a line on martin@idexfellows.com.