Risky Business

Risky Business


This post reflects my personal views only and is not intended to reflect the views of Change A View. My purpose is to memorialize my intent for personal accountability in the future. Note as of 4/20/2020: some of the .gif sources are no longer functional and have been removed but all else remains the same.

I. Was I just another shark in a suit?

A few years ago, I read Tyler Cowen’s The Complacent Class. In short, as I recall, it discusses a loss of American dynamism and diminishing optimism in society’s ability to innovate and change. It spoke to me; I’m in a profession that values precedent and discourages risk-taking. It is complacent.

Technology, particularly Internet technologies, have long gone part-and-parcel with optimism and free-thinking. There was a sense of liberation and inevitable utopia associated with Internet access. This was a place where eliminating barriers to communication would facilitate knowledge-sharing, writing, political engagement, and reduce authoritarian attempts to stop people from learning other, potentially better, ways to live their lives and run their countries.

Fast forward to 2019. We now know through shared experience that a “good” Internet is not self-executing. It takes work. We’re aware that we can build seemingly neutral tools that allow bad actors to do terrible and destructive things, ranging from individualized harassment to genocide and terrorism. We also know that the design of platforms and applications can unintentionally steer people in a direction or box them into a limited set of responses and actions. Likewise, we know that diminishing barriers to entry for communication allows lies and propaganda to flourish where we wanted truth to govern.

The response of popular platforms to these problems have been found wanting. They are complacent.

II. Mission Impossible

I’m not complacent. I’m excited. I’m taking a risk.

There’s a litany of writing about the ills of social media and other “discussion” platforms. Let’s level set: they are not discussion platforms. They are advertising platforms. The problems of other platforms are another post for another time.  I want to outline my personal take on Change A View’s vision.

I’ll do this by focusing on the questions our team has been asking ourselves. This is because I think they’re demonstrable of our approach and goals, and that people who do not ask the right questions will solve the wrong problems.

I am doing this for three reasons:

  1. Free societies require people who think critically. I think good conversations are important to the civic health of free societies and even though my political views have changed as I’ve grown, this is has been one of my bedrock first principles. If I can make it my role to create a product centered around facilitating those conversations, that’s what I want to do. I don’t think that this is something being thought about at all in the upper echelons of most technology companies and, while we will in no way be the sole purveyor of good conversations, I do think we can be a model for how to have a thoughtful and deliberate impact on broad swaths of people and institutions.
  2. We have an opportunity to be creative and thoughtful with design. I originally moderated Change My View because I felt I could do #1 on that subreddit. For a while, I could, but as you’ll see from the items below, ultimately our ambitions outgrew the tools Reddit provided. This is not a knock on Reddit. I still use Reddit. I really like Reddit sans some legitimate criticisms about how they deal with people I believe to be dangerous (another discussion for another time.) Change My View is still the best of Reddit, and its moderation team continuously raises the bar for professionalism despite being entirely volunteer. It’s simply a fact that Reddit is designed for one thing and we are designing for another. I wrote about the influence of design choices here, just as another data point that this isn’t something we conjured out of the ether.
  3. I believe in it. This means something to me. By far, the most critical comment we receive is: you’re doing this for the money 1. I helped create this because I find all of these questions important and fascinating. Yes, I would love for it to be my full time job to help answer them, but one of the risks I took in contributing to this project is that it might not succeed and therefore might not make money. Nevertheless, I earnestly think people want this experience, so I’m willing to take that risk.

III. Edge of Tomorrow

Here are some key questions we’re grappling with:


  1. What design choices must we make to ensure that everyone feels safe to participate on our site? In what ways could someone manipulate or misuse a feature in a way we hadn’t intended but could have foreseen?
  2. What is the online experience for people who have a different background from us, and to what extent should that inform how we design the site?
  3. How can we incorporate this into hiring decisions without unduly burdening women and people of color with the responsibility of mindfulness and awareness?


  1. How do we moderate at scale? How do we utilize the necessary tools of machine learning while still keeping humans-in-the-loop for grey areas?
  2. Where we do use AI and machine learning, how can we account for the biases often found in those tools? If we don’t develop them ourselves, how can we ensure that we’re cognizant of and accountable for those tools we adopt?
  3. What do we do to ensure care for the mental and physical well-being of these moderators, many of whom will be exposed to problematic, challenging, and frightening claims?

Rules and Good Faith Participation

  1. In what way(s) can we create a rule structure that incentivizes productive conversation without chilling speech?
  2. What constitutes “due process” under our moderation system such that people feel that removals are fair, consistently applied, and subject to appeal? How will this inevitably change as we grow and need to manage more posts and comments?
  3. People very rarely want benevolent views changed. How do we create a space where people can have a problematic view challenged while not creating a safe space for harmful advocacy or normalizing that harm against others?
  4. How do we maintain transparency without revealing details about people using the site and/or moderating the site?


  1. What metrics matter most to us? Number of views changed? The extent to which a view has been changed? How many people post a view to be changed?
  2. With whom can we partner to make other online experiences better — encyclopedias, periodicals, newspapers, schools, businesses, governments
  3. How do we answer all of the above in vastly different social contexts, such as countries that would hurt their citizens for expressing a particular view, or inadvertently exposing they’re part of a targeted and vulnerable class of people?

There are more.

I don’t have all the answers to these questions. I know that they have real world applications of significant gravity, and they don’t exist purely for me to fiddle with and think about. They nevertheless spur a desire to figure out some solutions to these problems, even if it only moves the needle a bit.

That’s why I raised Cowen’s Complacent Class in the beginning. I could see a world where a similar opportunity presented itself and I didn’t leap at the chance. There are a lot of ways this could go wrong, or I could end up becoming part of the problems I want to solve. These are risks, but they are all risks I think are worth taking.

As a founder, I think it’s important to be candid about where I am personally starting. I am, as you’ve probably guessed, open-minded to the possibility that some of my hypotheses are wrong, but that is the point: I am building Change A View to test these assumptions. If I’m off, then I’ll owe the world a delta.

Notes all Tom Cruise movie references occurred organically as I struggled to think of headings. I don’t have any aspirations to climb buildings or do risky business stuff. Well, maybe buildings, if supervised by professionals and a trampoline is involved. Additionally, I can’t recall where I found this image of She-Hulk, but I liked it and will be happy to provide credit if someone credibly claims it or can point me in the right direction. Finally, this will eventually be on ChangeAView.com’s main site as part of a founder introduction series, at which point the disclaimer about personal views will be removed. Until then, this is just me.

1 On Monetization. I’ve seen, multiple times now, claims that CAV is, in some nefarious or underhanded way, making money off of other peoples’ content. I don’t want this post to be a devoted rebuttal, so I’ll leave my rough position at this: this isn’t about “taking” one’s content. All subreddits uses other peoples’ content. They are part of a content aggregation site leveraged for targeted advertising. Yes, when a person agrees to our Terms of Service we could (for example) quote that in a blog post or something similar, but otherwise these posts net us zero dollars. In other words, if one is concerned about people making money off of their views, then I would highly recommend leaving any comment-based website. If you looked at most pages with a comment feature, they likely make the similar claim that, by participating, the website has some license to use your words. In our Terms of Service, we worked diligently to provide plain English summaries such that people could make an informed decision about whether they agreed or not. The “legalese” itself attempts for each sentence to be < 30 words to avoid confusion.