I was wrong. Not about social media, the low effort posting problem is real, and removing corporate incentives doesn’t fix it. But I drew the wrong conclusion from that diagnosis. I walked away from decentralized protocols entirely, and I think that was a mistake.

Let me explain what changed my mind.

Recently I’ve been spending time with nostr, a stupidly simple protocol for decentralized communication. And I mean simple in the most generous sense: it is almost shockingly minimal. There are no servers to trust, no central registry of identities, no algorithmic feed you can’t escape. You are a cryptographic key pair. Your posts are signed events broadcast to relays. Anyone can run a relay. Anyone can build a client. That’s more or less the whole thing.

The web didn’t solve human nature either. People used it to spread gossip and misinformation from day one. We built on it anyway.

When I first encountered this, my instinct was familiar: here we go again. Another decentralization experiment that doesn’t address the real problem. And it’s true, nostr doesn’t fix the low-effort nature of social posting. You can still fire off a half-baked take in thirty seconds. The protocol is indifferent to quality.

But I’ve been thinking about the wrong question. The question isn’t whether nostr solves the social media problem. The question is whether open, decentralized protocols for communication are worth participating in as they develop, even if individual apps built on top of them are flawed.

Think about what the web looked like in 1995. It was a mess. Most websites were embarrassing. The tools were primitive. The content was thin. A reasonable person could have looked at early Yahoo and Geocities pages and concluded: this encourages low-effort publishing, anyone can post anything, there’s no quality control, I’ll stick with books. They would have been right about all of that. They would also have completely missed what was actually being built.

What was being built wasn’t any particular website. It was a layer of open infrastructure, HTTP, HTML, URLs, that no single company owned. That foundation is why the web could accumulate thirty years of value without any one entity controlling the toll booth. The bad content on early websites was a distraction from the thing that actually mattered: the protocol itself.

Nostr is a protocol. Not an app, not a company, not a community. The apps built on it range from Twitter clones to encrypted messaging to long-form blogging to marketplaces to git repositories. The protocol doesn’t care what you build. It provides identity and message-passing with cryptographic integrity and no central authority. That’s the foundation. What gets built on top is the open question.

I used to think my job was to find the perfect platform, one that would make me a better, more thoughtful communicator by design. I was looking for a product to solve a personal discipline problem. Nostr doesn’t do that. But neither does the web, and I don’t hold that against it.

What the open web gave me was the ability to publish this blog, on my own domain, using open standards, in a way that no company can take away or bury in an algorithm. Nostr offers something analogous for social communication: your identity travels with you, your follows are portable, no relay can silence you globally, no company can shut down the protocol. These aren’t small things. They’re the conditions under which anything interesting can be built without asking permission.

There’s a window for this. Open protocols succeed when they gain enough adoption that the network effects lock in the openness, when walking away from them becomes genuinely costly. The web hit that inflection point. Email hit it. If nostr or something like it hits it, we get a communication layer that no one owns. If it doesn’t, we continue with what we have, which is a handful of corporations mediating nearly all human digital communication.

I think that’s worth participating in. Not because the current apps are great, many of them aren’t, and my original criticisms apply to how people use them. But because participating in protocol development while it’s still early is how you influence what gets built. The people who shaped the early web weren’t just users. They were building, experimenting, arguing about standards, making tools.

I’m not suggesting everyone rush to nostr. Most people have no reason to. But I’m a person who cares about how the infrastructure of communication gets built, and I’ve decided it’s inconsistent of me to stand on the sidelines of the most interesting experiment in open communication protocols happening right now, while grumbling about the apps built on closed ones.

I’ll keep writing on my website. Long-form still demands effort in a way that short-form doesn’t, and nothing about nostr changes that. But I’ve stopped treating protocol skepticism as sophistication. Sometimes it’s just caution dressed up as wisdom.