A Social Protocols OS

My wife and I often talk about the difficulty of making friends as adults. There’s lots of articles online about this so I won’t rehash them — what’s been on my mind recently is that the root of the problem is broader than it appears, and understanding that root leads to novel solutions.

The root issue is that we’ve collectively forgotten how to run complex social arrangements. Institutional decay is everywhere and our societal base of knowledge/skill to revitalize them have weakened.

In the past, making friends was a byproduct of participating in the larger social protocols of society. You meet people at church, civic groups, gardening clubs, etc. and gradually become friends and frequently hangout. These rich lower-level relationships strengthened and informed the higher-level institutions.

Because people participated in these social protocols from childhood, they were more skilled at running them. To give some literary examples, Mark Twain’s “Cannibalism in the Cars” narrates how a group of snowbound train passengers instantly enact comprehensive parliamentary procedures to vote on who gets eaten. Twain clearly assumes that any educated person in the story and reading the story would be familiar with parliamentary rules. The series of teenage novels “Mad Scientist Club” written in the 60s in the US has a group of teenager friends who decide to create a scientist club and immediately elect officials and a treasurer for shared expenses w/o any preamble to explain how they all work. I suspect not even 1 in 10 people alive today could organize and run a club without reading a book about it first.

This isn’t a new thesis — the book “Bowling Alone”, released in 2000, extensively documented the decline of social and civic involvement from the 1970s on — pinning the blame largely on TV, that it proved more interesting than hanging out with people.

Social media is furthering this trend of digital leisure activities supplanting IRL socializing and civic involvement.

I’m not arguing against this trend per se — TV and social media are very fun — I’m perfectly happy staying home most nights of the week — but I’m convinced we’ll always need close IRL friends & varied social interactions to be happy & generative throughout life. Many of the richest experiences we can have are in deep complex, creative relationships with others.

But we’re not going to stop watching TV or scrolling through feeds. So how can our society bootstrap the individual skills and group social capital necessary for the vast majority of people to have great social experiences and to keep solving all the problems reality keeps throwing at us? Build a sort of Strategic Social Capital Reserve?

Many of us in the Web 1.0 and early 2.0 eras expected the internet to help with this. And while I love Twitter, etc. and have found many friends through them — they have a big weakness — social networking platforms have little incentive to help us form friendships or to support strong IRL communities. They’re ad-driven and ads work best when we’re glued to our screens consuming the best content from around the globe. They’d much rather us quote tweeting each other in dark rooms in our underwear than enjoying a relaxing dinner out with friends.

Cory Doctorow wrote recently about how and why social networks steadily get less aligned with our interests — Tiktok’s enshittification.

What if we had social networks that are actually incentivized to help us form complex friendships and groups that work well together over time?

There’s a lot of groups exploring this space right now.

New social networks include Farcaster, Bluesky, Nostr, and Lens. These are all based on decentralized ideas and rely on Blockchain technologies.

Decentralized autonomous organizations (DAOs) are exploring how to safely collaborate with internet strangers without centralized leadership.

“Summer of Protocols” is spinning up a summer research program to take a broad look at protocols. Their “whitepaper” The Unreasonable Sufficiency of Protocols and reading list are an excellent deep dive into protocols (if this is interesting, the application deadline is March 21)

My mind keeps going to what seems like a missing hole in all this — how do we make it easier to design, share, and run complex social protocols? We have low-level social primitives like messaging and group “containers” like Facebook Groups, Reddit, and Discord but nothing it seems that helps these level up to more sophisticated behaviors.

The world has changed a great deal in the last few decades and many of our old protocols don’t seem to work as well anymore. We need new easy to learn and run protocols (probably mostly digital) that help reduce friction to hanging out and working together.

The SaaS software explosion might provide a lead — it’s been 20 years of taking business processes run in webware (often topped with gray hair) and porting them to web apps. This improves business efficiencies as they encapsulate many of the hard parts to business processes so companies operate more efficiently.

Digitally mediated social protocols make a lot of sense to me. And perhaps my brain has been addled by too much time writing open source and frameworks but it seems like there should be a protocol OS and framework for designing and running social protocols which then can be invoked by groups.

Want to run a club? Invoke the club protocol and you get yearly elections of officers, dues collection, etc. Want to run a book group? Pick one of five popular maintained book group protocols off the shelf and get going. Want to do a monthly potluck dinner with a few other families? Grab it and it’ll help you arrange who’s hosting, the date, the time, dish assignments, etc.

This Social Protocol OS can be a layer that runs on top of social networks and messaging apps and arbitrary protocol programs can be written to run on top of it.

The common theme of social protocols is they help facilitate the coordination that must happen so that it’s easy to get to the fun part of hanging out and working on stuff together.

Alfred Whitehead said “Civilization advances by extending the number of important operations we can perform without thinking about them”. What if many social protocols that are currently very difficult to do became almost trivial? E.g. I’ve never run a conference but those that have say it’s shockingly hard. What if it was approachable by a normal non-operations genius? It seems like the world should have far more 10-100 person conferences than it does right now.

There’s an immense depth of social protocols from societies around the world that we could lean on for revitalizing society after the upheavals of the past decades. An internet-mediated fusion society. We regularly eat and cook dishes from all over the world. Why can’t we mix and match from the best social protocols from around the world and across history?

If a DAO figures out a more effective way of doing things — why can’t they release that as an open source protocol? Or PR improvements to a library. We could encode the best collaboration/decision-making protocols in open source code and iterate on them together. A composable library of social protocols we can use together to create familiar and novel communities.

I use a running and weight training app which reliably takes me from terrible shape to very fast and strong in 6 months. Could we do the same for groups? Scale them from simple low-trust interactions to complex high-trust interactions over a year or two?

Or is this too cyperpunk-ish and cold and unworkable?

My main question now is if there’s a there there. This is a complex space with a lot of history. If you’re interested or have thoughts — please reach out, I’d love to chat. Also a concrete use case or two would be idle for tinkering with the idea e.g. you’d like a Discord bot to coordinate your group doing something.

Thanks to Shannon Soper, Kerry Soper, Sam Bhagwat, and Venkatesh Rao for reading drafts of this post

Subscribe to get updated on new posts!