What do we do with sparkly people?
On building environments that encourage intelligence, curiosity, and agency.
Created with DALL-E 3. Prompt: “a small number of human brains connected by glowing lines”
You should read at least the intro and TL;DR of Logan Graves’ essay on social infrastructure to grok this post. A lot here is directly inspired by Logan’s post and conversations with him on this, but shouldn’t be construed to be representative of his ideas or beliefs.
Notes from future me:
“sparkly” was a dominant part of my vocabulary until I read this. Now I try to be much more mindful about using it.
I wrote this at a time when I was very frustrated by the train wrecks that were the clubs I was in and my own inability to run a good club. Looking back at this, I think I had the right idea with a lot of my “solutions”, but I also think I should’ve toned down the level of centralization and top-down control I was advocating for.
I don’t think I would be as strong of an advocate for rewards anymore. They often don’t work as well as you would expect, and even if they do, they probably cultivate the wrong values. Also, Goodhart’s Law.
01. Recap
I use social infrastructure to refer to a process or institution that connects people in ways they otherwise wouldn’t be connected.
Specifically, people like Logan and I want to build scalable social infrastructure that connects and enables more sparkly people (people with intelligence + curiosity + agency). For convenience, I’ll refer to “sparkly social infrastructure” as SSI for the rest of this post.
Importantly, (at least for me) SSI doesn’t just look like a social network for highly ambitious and curious teenagers.1 I want to build something that would’ve delighted Vihaan2022. There were lifestyles that I didn’t know I could live and identities I didn’t know I could develop because I never felt like I had permission to not be normal. SSI should be something that explicitly gives permission to all the Vihaan2022s.
In an ideal world, education systems would be better at filling this role for the average person,2 but in the absence of expansive education reform, well-designed social infrastructure can hopefully be a good approximation.
Here, I want to move closer to a concrete outline of what I would want SSI to look like in a world without major funding or scale bottlenecks. What do you do at SSI? What does it feel like to be a member of SSI? What does the organization itself look like?
Over the next 5 years, and especially after high school, I want to experiment with variations of this outline to grok (a) what makes communities work, (b) what makes people more sparkly, and (c) how feasible a scaled version of (answer to b) even is.
02. Analyzing Hack Club
The goal of Hack Club is to help you become a hacker. We want a space at every school where people are making interesting things with code, every week. Schools don’t provide that, so we’re creating it in every school to make building things accessible to everyone.
Hack Club is a great case study for SSI because it broadly tries to do something similar. If you tweaked it to not explicitly be about programming, this is what an SSI mission statement would read like. I want to analyze Hack Club through the lens of someone interested in building something similar. This is somewhere between an outside and an inside perspective and I know very little of the Hack Club lore, so I’m definitely going to get some things egregiously wrong, but I’ve found the analysis to be instrumentally useful.
The Good
At least on a surface level, Hack Club has done very well in the short amount of time it’s been around. It has a wide reach, amazing yearly summer and winter events, several cool incentive systems, and it’s practically an incubator for tons of impressive collaborative teen projects. If you’re a proto-hacker and you genuinely immerse yourself in everything Hack Club has to offer you, it will probably be a significant high-growth experience overall.
Reasons for success:
Most of the people with positions of authority are young, great hackers in their own right, and clearly very invested in the mission.
The Hack Club Slack is probably the most important asset Hack Club has for fulfilling their mission and Hack Club leadership puts in a lot of work to keep it comfortable and alive.
Being a lonely teen hacker is something that a lot of now-wealthy people have experience with. With good connections, that probably helps with funding.
The Bad
I want to make it extremely clear that I love Hack Club and I have a lot of respect for the people at Hack Club HQ. Any criticism comes from goodwill.
From personal experience, hearing from friends, and an understanding of what the average high school club looks like, I think it’s pretty safe to assume that a good chunk (my guesstimate is > 50%, but that may be too pessimistic) of Hack Club chapters are dysfunctional and have marginal impacts in terms of helping their members become hackers.
At the average chapter, there are probably a few members who show consistently, some members who show sporadically, and a majority of members who barely show at all. If the leadership ever genuinely cared, they’re probably jaded and don’t try as much anymore. This is all assuming that there was even enough interest at the school to get decent membership in the first place.
Unless you’re lucky and happen to be part of a well-functioning Hack Club, the only way to immerse yourself is by being engaged with the online community. The issue is that many people probably don’t have/want to take the time to keep up with a community of 30,000 people, or just forget to ever log into the Slack again. For me, the whole experience of trying to find channels I might enjoy, understanding the different communities’ norms and mores, and staying on top of conversations was just really high friction.
The people most likely to make it through and become engaged members of the community are those already highly predisposed to be hackers. That’s great, but it makes Hack Club more of a community for hackers than an environment that helps them become hackers.
What goes wrong there? I think Hack Club makes a bunch of critical errors rooted in optimism.3 I think it’s too optimistic about club leaders genuinely caring in the first place and too optimistic about clubs operating well by default even with leaders who do care. I think it’s too optimistic about the online community’s ability to substitute for the clubs for most people. It’s too optimistic about the impact that most “you ship we ships” have on people who are on the fence (Arcade is a notable and lovely exception).
Solutions
The Slack is great, but I want to preserve small groups (ideally <50). Purely based on personal experience, I believe that they’re better than big communities at transmitting relatively uncommon, hard-to-transmit, and demanding (in that it’s higher effort to be a hacker vs. a banana lover) traits. Either way, you should be able to get most of the Hack Club experience without being tuned into the Slack, and I think that’s what Hack Club leadership originally intended.
I see a few ways to make intimate groups the center of gravity for Hack Club again.
Solution 1: Hack Club makes the average chapter better.
We’re asking teens—many of whom don’t care much or have tons of other things on their plates—to engineer a culture that not only provides a space for hacking but also actively makes hackers of those who would otherwise be on the fence. I know how hard it is to build a successful culture and keep a club from falling apart even when you really do care. Hack Club should err on the side of overcorrecting.
Jams and workshops are great, but they’re a small part of what’s needed to make the average chapter better. Where is Hack Club’s version of Socratica’s toolbox? If it does exist, as a club leader who used to be relatively involved in the community, it’s a failure on Hack Club’s part that I don’t know about it.
I would have Hack Club make a toolbox++. While not strictly required, there should be some sort of positive incentive for chapter leaders to read/watch this thing through that feels more than worth the time cost. Maybe there’s a basic quiz to prove that you at least took the time to skim it.
There should be additional incentives to keep Hack Club HQ in the loop about how the club’s going, what steps you’re taking to make your existing members better hackers, and what steps you’re taking to reach and encourage people on the fence. Hack Club HQ should organize more events and meetings for leaders to discuss their clubs.
I see a few positive side effects to all of this.
Hack Club HQ could also serve as a bridge between clubs. They could recognize when two clubs near each other have attendance issues that could be fixed by teaming up or want to work on similar projects.
If done well, it can almost sanctify the idea of being a chapter leader to the leaders themselves. It can help them see it as an important role with intrinsic worth outside of college applications.
Solution 2: Hack Club creates intimate groups online.
If the first solution isn’t scalable or doesn’t help much, Hack Club can design intimate groups within the Slack. People in the online community would be encouraged to join groups based on timezone and availability that are just online versions of the clubs.
These “virtual clubs” would be quality-controlled by Hack Club with leaders who seem to care and have been effectively taught how to engineer communities, and members would be incentivized to attend meetings through something like Arcade. This would ideally overcome the low-attendance issue that a lot of Hack Clubs have and provide more flexibility. You could join the same group each week, or hop around based on your availability. Some people will want to stick to one meeting per week, while others might join meetings every day.
03. So what do we do with the people at SSI?
I don’t feel like my relationship to sparkliness is best captured by 1-1s, competitions, and student-run lectures. They’re great, but they give off the vibe of being isolated, relatively high-friction, and something you go outside of your normal day to do. I think I thrived best in just having a specific and consistent space where I felt comfortable being sparkly and I could see others being sparkly, in an environment that felt deeply personal and close to my everyday life.
I think the center of gravity of the SSI experience should be some sort of default state where you’re able to just be sparkly and be around sparkly people in a relatively low-friction way. Specifically, I’m envisioning Socratica-inspired sessions. Every week, you can dedicate time to learn/make/do something surrounded by other people (unlike Socratica, I don’t think the emphasis should just be on making), and you hopefully continue doing that outside of the weekly sessions. Outside of these sessions, there should be consistent “special events” like the essay contests, speaker events, participant-run workshops, and hackathons.
To control quality, I think the ideal source of activity should be in city/region chapters (contra to how the ideal source of activity in Hack Club is in individual school chapters). School chapters will still exist, but only as a glorified recruitment mechanism. From there, it looks a lot like my second solution for Hack Club.
There would be several online sessions set up for each week with quality control by the regional chapter. The regional chapter is be responsible for organizing the recurring special events, but provide resources to make it easy for people to organize their own (like Hack Club with HCB). There is some sort of online community that you could choose to engage with but isn’t strictly an important part of the SSI experience. I’m not entirely sure how regional chapters would be selected and run (current members, volunteer former members, paid staff), but that would clearly be a critical early decision for this model of SSI to make.
Some failure modes of this whole plan:
Small groups aren’t much better than larger communities and turn out to be lower friction
It’s harder than I thought to engineer small communities even with quality-controlled leaders.
People just aren’t interested in these sessions.
If any of these are true, then the SSI experience should be built around the “special events” and the online community instead.
If you have thoughts on any of this (especially engineering small communities around specific trait(s) and large communities vs. small groups), I’m always open at vihaansondhi07@gmail or @vhaangogh on Discord.
Although that is valuable in its own right, and just a few months ago I was really enthusiastic about building it. I probably still do want to build a prototype of it in the near future (even if it would just be useful for getting better at frontend programming and design).
The American education system surely isn’t anywhere near there yet, and I’m making the lazy assumption that the vast majority of others aren’t as well.
If you read to the end, you would probably be right to argue that I’m also too optimistic. I’m unreasonably bullish on the impact that small groups have, especially because there were lots of contingent factors about my own experiences with successful small groups.