Express yourself without fear

Creating a design solution for mediation

Growing up, I was quiet, fearful of what others thought of me. At school, I aced written tests — math, history, science, English. Anything that didn’t involve speaking. When talking was required, I failed. I received F’s in participation. Nobody could hear me during class presentations. Oddly, I would mouth words during chorus rather than singing. “Why are you so quiet?” was the most common question written in my yearbook.

But when the digital arrived in the form of a Macintosh Performa (and 1200 bd modem), my life changed. I could type without having to speak. I connected online to many strangers. Some were great debate partners. Some would be real-life friends. Some were love interests. I learned how to navigate the world. I learned to create a compelling rhetoric. I could find my own voice through words.

What I remember the most during that time was how the digital world was inviting and innocent. In college, social networks and online blogging communities provide a place to experiment with identities. I joined each one, eager to connect with others. I absorbed information. I made a short-lived podcast. I wrote about my roommate issues, perspectives of work, and gossipy thoughts about friends.

I found the person who I wanted to be — melding the offline and the online.

In person, conversations are inherently transient. But online, language is permanent and sometimes brutal — with its ability to re-read words over and over again. We could experiment — telling jokes at an open mic or sing at a campfire.

But today, the world is different. We are all better connected than ever — we find long lost friends and remember what we discussed. Social media today, shaped heavily by engagement loops, encouraged all of us to be perfect. This permanence and convenience have allowed us to return to the era before the digital world. It’s easy to judge quickly.

As a byproduct, we cripple from the potential of public scrutiny or even worse shame. Yes, some do fearlessly speak. But there are so many of us who have chosen to stay silent.

For the past year, I had pondered this topic, but never quite fleshed it out to reveal its core needs.

What kind of safe space can be built?

The Opportunity

Having heard of Reverse Hackathon from Sandra Sopol, an organizer of the event, I signed up for a Saturday of designing, prepared to be open to other ideas. During the team matching activity, I stood among the “relationships” group and mentioned my thoughts above. “Shame is too prevalent,” I said. “I am interested in finding a way to build a safe space.”

To my surprise, others were interested. We formed a team — Hannah Lee, Dani Reis, Liz Carey, and myself. We found our space in Room 306 in the Center of Integral Studies. Our background and skills spanned from design research to marketing to user experience design.

Process

With four hours to “hack”, we dove into our work. First, we went by a great philosophy used in product development — let’s figure out the right problem rather than figure out how to solve the problem right.

As a fan of the Google Ventures design sprint process, I was thrilled that we incorporated the method to elicit ideas to arrive at the core of the problem and iterate.

Dani even happened to have a mini version of the timer, recommended by the GV Design Sprint. I think that I might get one myself.

Goal Setting

First, it was important to agree on a common goal. Together over five minutes, we brainstormed goals on post-its. We clustered and discussed over 10 post-its about how to address communication about tense topics and self-expression.

User Definition

Then, we discussed the potential users.

At first, we headed toward a younger population. We believed that teenagers and young adults — people who are still figuring out who they are — may benefit from safe spaces as they define their identity. Hannah pointed out that there may be Children’s Online Privacy Protection Act issues, which would limit the flexibility for users under 18. Yet, Liz highlighted how such issues can span all ages. An older person may struggle as much as a younger person. We decided to be age-agnostic.

In particular, we settled on marginalized users from introverts to people of color to those from the trans and queer community. Although our target user later shifted, we still believe that our solution may empower these users the most, especially in conversations where they are marginalized.

Ideation

We defined a problem space — a community with an onboarding vetting process, an in-person event, and ongoing engagement. We decided to focus — particularly at the meeting point.

We had an initial idea of a “get coffee” button. If two people online got into an escalating tense discourse, one party could hit this button to trigger a request to meet in person. We believed that by meeting in person, people would recoup the compassion lost in digital conversations and still dive into a fruitful conversation where both parties could walk away not necessarily agreeing, but appreciating and respecting the other side.

At this point, two hours or halfway through the hack had passed. I put out a call for experts on Slack. We wanted a gut check to confirm our direction. In two different shifts, four experts with experience in psychology and coaching arrived to provide feedback and expertise. We walked through our problem space and our initial idea.

One expert pointed out the disparate intentions:

1. Allow people to express themselves freely (e.g. many struggle to share their art, because the fear of others judging harshly and a feeling not good enough)
2. Resolve an escalating discourse so that both parties could be heard (e.g. the example above where parties dive into a disagreement that polarizes rather than provides information).

The disagreeing parties may not want to meet in person. Their mental frames of a meeting may be differ. Unfamiliar, dangerous situations could result. One party may not be ready to have a conversation. Logistics become a barrier to meeting — deciding on a place and a time convenient to both parties (assuming that the parties are local) could not scale.

I was curious about how we could balance values and morality in our experience. For instance, white nationalist groups arise from communities where everyone bonds together, because they have similar beliefs. How could a community not veer toward extremism? An expert suggested that those principles of openness and integrity need to be integrated into the community. Moreover, that a moderator of the community would need to embody those values.

We decided that the most effective solution should be built on top of existing platforms. Building a new platform would require significant effort to amass users. Current platforms work. We considered Reddit, Wikipedia, Facebook, Twitter, and Craigslist. Yet, we decided that our solution would be device-agnostic.

Based on the feedback, we resolved to iterate and arrive at a tangible solution.

The Pitch

Our mission is to create a safe space for online communities. Under the right conditions and tools, we believe that an online community can thrive. Users may not agree with each other, but they can respect each other.

Problem Statement

How might we build a space so that users feel safe to express themselves without the fear of retribution?

Do you remember when we could have a conversation and feel that we heard both sides? In person, because of the cognitive frame — that we have arrived together by our own free will through friendship or similar, we have chosen to converse. Yet with social media, that kind of frame no longer exists.

Current Landscape

In today’s world of communication, words are permanent and easily disseminated. This leads to polarization — hate groups arise and conversations halt. With the breakdown of communication, users lack the education or encouragement for supportive language. As a result, users are unable to build bridges and human connection around disagreements.

Target User

We believe that the target user wishes to engage in a conversation with respect, tolerance, and acceptance. Through the experience, users could create authentic and organic conversations to grow and stretch their comfort zone.

These users would already be on a platform (e.g. Facebook). They seek inclusivity that could include diverse political leanings, gender, race, economic status, abilities, age, religion etc.

Opportunity

In today’s environment, today want a safe space where they need not fear repercussions if they have a differing opinion. This space would provide a medium to engage in a conversation that enables both parties to be objective and maintain an “agree to disagree” mentality.

Many online users feel paralyzed to engage in discussions to due to fear, lack of confidence, and trust.

Solution

Introducing…

An aid using machine learning and natural language processing for tough conversations that builds skills for communication and strengthens connections between parties.

Scenario

Imagine that Cheryl, our target user, is having a conversation with her Uncle Bob on Facebook. She writes about a current dilemma. To her surprise, Bob says something harsh. Cheryl reacts immediately to the direct comments while not sensing what Bob intends. As she does so, she sees PandaBot surface. In this situation, PandaBot suggests inviting Bob to a meditated conversation.

Cheryl agrees.

Pandabot first invites both parties to answer a series of questions. Pandabot will not reveal the answers until both parties have answered the question. This way the conversation now becomes balanced and unbiased.

The intention is to allow each party to reflect on the conversation while considering their own feelings and the feelings of the opposing party. Pandabot aims to level set the conversation so that both parties will feel that they have been heard.

After a few rounds, Pandabot assesses the situation. For instance, Pandabot asks each party to rate their feelings of the conversation. Using that response, Pandabot makes a conclusion about the next steps—restart the conversation, allow time to pass between now, or continue additional rounds.

Pandaboth helps people communicate through mediation and sharing listening / reflecting methods.

Results

During a “science fair” round of 30+ teams, we pitched to a round of experts. Our team got to the top ten. We had an opportunity pitch to a panel of judges and our fellow hackers.

The experience was amazing as was meeting and working with such brilliant colleagues. With more time, we could iterate on our idea and build it into something tangible that could benefit the world. Due to the time constraints, we arrived a solution that only supported a portion of the initial problem space. If we had a whole weekend, I imagine that we would map out a potential journey, a roadmap, and a stronger MVP of Pandabot.

As a fellow hacker mentioned, it’s not that any platform — Facebook, Twitter, Snapchat, Instagram — are inherently evil. They are useful. They were built by us for a purpose: to connect with others. They have the pluses — where we can connect with those who we don’t see that often and remain in each other lives. Yet, because we have designed them, we achieved our goal of connection, perhaps even over-connected. We are only human. How can we make this experience more human again?