Dangerous protocols

After the early crypto boom of 2017, I spent a few years dabbling at its edges. I was deep into open source world then, and curious about crypto, which seemed to be open source culture embodied in an economy.

It was fun to be surrounded by people who shared my intellectual interests, and I had a lot of questions for everyone I met. One of the earliest questions I remember asking someone was about the governance of distributed file hosting. Imagine a file that’s not held in one single place, but split into a thousand pieces, which are held across many different providers. No one provider can take down the file. This is great for many reasons – security, reliability – but immediately raised other questions for me.

“So, say a vindictive person uses this protocol to publish compromising photos of their ex. Who does the ex contact to take down the photos?”

“Well,” the person I spoke to paused thoughtfully. “I guess there would be a governance council of all the hosting providers, and they’d have to establish a policy on how to proceed.”

I made a face. “You’re asking all the right questions,” he added. “These are the things we’ll have to figure out.” But the years passed, and I never got a satisfying answer from anyone.

For advocates of the decentralized web, protocols are often viewed as a good thing – a liberating alternative to, say, platforms like Meta and Twitter. But decentralization seems to be at odds with the historical purpose of protocols, which is to simplify communication. Most people would rather have one identifiable adversary to grapple with, instead of a thousand faceless ones with middling levels of motivation: or worse, not know who their adversary is at all.

As the culture wars ballooned, I realized that protocolization, and its relentless tendency towards simplification, had much bigger implications. It seemed that everything – how we thought, how we acted, who we dated, the opinions we held, even our intellectual pursuits – had become protocolized, and always with a tinge of urgency. Understanding protocols, then, was a way to understand our present cultural moment.

In the 1930s, the rise of bureaucracy helped us simplify the information boom that came out of the Industrial Age (Protocolization 1.0). Similarly, today’s crisis mindset is an attempt to simplify a boom of opinions that have proliferated in the Digital Age (Protocolization 2.0). LLMs are the expected output of this era: an attempt to compress human creativity into something predictable.

For the Summer of Protocols research group1 last summer, funded by the Ethereum Foundation, I decided to explore protocols as systems of control. The result was an essay on dangerous protocols: how protocols usurp our agency, and how we can subvert them.

Read it now

Just as bureaucracy – despite its downsides – propelled humanity forward by making it easier to organize at scale, I think the crisis mindset emerged as a way to help people prioritize their attention. But if everything is running on autopilot now, it’s especially critical for us to swim into the machine and identify the kernel of that-which-makes-us-human that enables us to steer the ship.

I’m still trying to figure out how to navigate and make the most of it as a lone brain floating through this newly protocolized world. If you feel similarly, you might enjoy this piece. Let me know what you think!