Reimagine the Internet Day 1: Pioneering Alternative Models for Community on the Internet

Posted 5/10/2021

This week I’m attending the Reimagine the Internet mini-conference, a small and mostly academic discussion about decentralization from a corporate controlled Internet to realize a more socially positive network. This post is a collection of my notes from the first day of talks. The full recording is available on the Reimagine the Internet site.

Great summary quote from Ethan Zuckerman: “It’s not too late to fix things, but we have to stop fixing what we have now and imagine what is possible.”

What We’re Rejecting

Large scale networks usually follow one of two models:

1. Surveillance and Advertising: Social media, generally any “free” services, which sell user data to advertisers for profit

2. Transactional Sellers and Buyers: Amazon, eBay, Patreon, any service that directly sells a product, or facilitates explicit financial transactions where the platform takes a cut

Large scale networks follow these models not only because they’re financial models that can scale, but because market forces demand scale. Public for-profit corporations are by definition growth-oriented, but there are stable non-growth models outside this dichotomy.

A great example is Wikipedia: they’re donation driven, but mostly sustained by small non-ideological donations. People find Wikipedia valuable, for their homework or quick-references for questions in any context, and they’re willing to throw a few dollars the encyclopedia’s way to keep that product available to them. In this sense Wikipedia is extremely product-centric - their “growth” (mostly adding new languages and outreach in poorer, less Internet-dense countries) does not earn them profit, and is subsidized by providing a useful, slick product to their English-speaking base.

Small Communities

Facebook is ungovernable because it is not a community, it’s millions of communities bumping into each other with incompatible needs. Reaching consensus on content policy for three billion users is impossible, and a foolhardy goal.

Creating the largest community possible should rarely be the goal. Instead, we should create countless small communities, with their own content policies and standards of acceptable behavior. Users are in many such communities concurrently; we’re good at context-switching. Reddit and Discord are great examples of this: larger subreddits and discord servers are rife with abuse, but there are countless micro-communities with every imaginable focus that thrive and are generally respectful and well moderated.

Purpose-based networks are especially successful: Networks where users are all gathered for a specific shared purpose. Social networks are hard to design because many people will have diverging opinions on what constitutes “social behavior”. Wikipedia’s mission of “making an encyclopedia” is much more clear-cut. They have disagreements over policy, but everyone in the project is generally there for the same reason, and that agreement has made it easier to govern and scale.

The Fallacy of “Free and Open”

A lot of early Internet action-groups, including Wikipedia and the EFF, were based around a cyber-libertarian concept of “free and open communication”. Basically the ultimate expression of the “marketplace of ideas” - you can communicate with anyone on the Internet, regardless of race, nationality, sex, or age, because no one can see what you look like, and the Internet knows no borders. The liberatory potential of “radical openness” is outlined in the Declaration of Independence of Cyberspace and The Conscience of a Hacker (also known as The Hacker Manifesto), and is echoed in countless other messages from that era.

This vision of utopia has fallen short. Erasing race and sex and age, has (surprise surprise!) not eliminated bigotry, nor has it consistently led to a brilliant meeting of the minds. In particular, “free and open” amplifies existing power structures: the loudest, most agreed with, most online voices tend to get their way. This tends to mean mainstream communities with enough financial security to spend lots of time online.

Community Norms

Maybe one third of community behavior or less is governed by explicit rules; most behavior is governed by unstated norms of what constitutes “respect” or “acceptable behavior or content.” As an example, there’s no law preventing a student from standing in the middle of class and debating a professor over each point of their lecture, but this is generally unacceptable given the roles or student and educator, and is highly unusual.

When building platforms, we should keep in mind what rules and norms we expect our community to abide by, and design features with those norms in mind. Returning to the previous example, a web conferencing system for education may allow the lecturer to mute all students while they are presenting, requiring students to raise their hands before speaking. This reinforces social norms and makes it much more difficult to misbehave within the confines of the platform.

Reinforcing social norms goes a long way towards community self-governance, limiting the need for explicit moderation, and making a community more scalable and less chaotic and disrespectful.

Coexistence with the Corporate Net

Alternative networks are often presented in opposition to the corporate dominated “main” Internet, but this doesn’t mean that Twitter and Facebook need to burn for better communities to thrive. In fact, existing mainstream networks can be a valuable bootstrapping system to spread broad messages and find peers to join micro-communities. Alt-nets don’t need to be completely self-sustaining or discoverable at first.

Conclusion

A very exciting first day! While the talk was non-technical, I think bringing these ideas of small-scale, self-governance, norm-driven, mainstream-bootstrapping communities to peer-to-peer network design will be rewarding. Click here for my notes from day 2.