Hey Besties!!

The next couple of week's essays are inspired by deep dives into different spaces as I'm _potentially evaluating startup ideas. Why am I deciding to do this? Who the fuck knows.

This week's essay is inspired by community platforms and community, specifically the orbit model created by Orbit and this podcast from TKP

Community has been a hot buzzword for the past year. Given that the world has absolutely gone to shit and that everyone has been stuck at home, people have been yearning for people to connect with. Communities have sprung up in so many directions, from students looking to break into design and VC (yikes) to more wholesome groups like Reboot and rating dogs. As someone who's had the opportunity to be a part of a few groups and scale others, seeing the orbit model feels like such a dehumanized take on how we as humans should bond. That's not to say it's not important. We spent the past couple of weeks figuring out how to put KPI's on community (tldr: I still don't know what they should be) so being able to quantify how people are interacting and thinking together is important.

However, what's more significant than the pure numbers of community is the notion of connectivity. In an episode on The Knowledge Project with Noreena Hertz, she discusses the loneliness crisis. In its essence, technology has resulted in a net negative for how connected we feel towards people. Social media drives up anxiety and depression rates among teenagers, some people even quoting that they feel like they are unloved and did something wrong when their post on instagram gets only a couple likes. People are flocking to joining Slack groups at length. Humans aren't infinitely scalable. We don't have indefinite time, attention, and lord knows our memory sucks. Memorizing people's names is already a challenge. How are we supposed to remember people in over fifty communities?

The short answer is that we aren't. In a world that is overflowing with information, people have to be choosier with their time. What to spend it on and who to spend it on. That's why building communities at scale is difficult. This is for two main reasons.

The first is that in a sense, there are reverse network effects. As the community scales, there is less time spent meaningfully between each 1:1 interaction. I fundamentally believe that the relationships are driven from spending meaningful time between two people, that hanging out in a group doesn't necessarily mean that you're close to everyone. As such, the individuals in the group might spin out, or might not hang with the rest of the group.

The second, which is similar, is that there will always be niche groups that can be broken down within each community. Simply having a shared identity in "VC" (whatever the fuck that means) doesn't mean that you'll be interested in the same topics, come from the same background, or have the same values. As someone who values intersectional identity and reads a lot of diasporic novels, its become increasingly clear that having shared experiences in how you were raised has created some of the strongest bonds.

Short Aside because I like graphs LMFAO

If we want to formally describe it, we can say that the strength of the graph is dependent on how far the graph is from being complete, and the percentage of nodes that are connected to each other.

Obviously, this is not possible. If this were the case, everyone in the world would be connected. LinkedIn wouldn't exist and there wouldn't be niches in Twitter. As such, we're left with the idea of strongly connected subcomponents, which, by definition, are subgraphs in which it is possible to start at a vertex and reach any other vertex in the subgraph. Networks are typically seen this way. LinkedIn leverages this fact by telling you how many mutual connections you have.

But, if we're evaluating a community, one KPI to optimize is minimizing the number of strongly connected subcomponents that we have. By taking a poll of who knows who and how they feel about their friendships in the group, we can create a representation that we can algorithmically solve.

If you've made it here, thanks for listening to the rambles and best of luck with the week! If anything doesn't make sense, let me know! I'm trying to write more so I'll probably stumble a lot tbh.

Other stuff I'm up to:

• Going to NYC for two weeks at the end of July / start of August. If you're there, let's absolutely hang!
• Starting to build out my recipe menu for food and drinks that I like. If there's any recipes that you like, totally send them my way!
• Started reading Blockchain Chicken Farm, Red White and Royal Blue, and Black Swan. If you have a goodreads, would love to be friends! Add me here!