We Burned the Town Square Down. Now Everyone's Arguing About the Permits.

The Ruin Has Been There a While

Picture the moment someone realizes the party ended hours ago. Not when the music stopped — that was gradual, almost imperceptible — but the moment they look around and understand that the people they came with are gone, the lights are unflattering, and they've been talking to strangers out of habit rather than desire. That's roughly where we are with social media. The writer at Ars Technica framing this as "RIP" is generous. The obituary is years overdue. What's interesting is that someone is finally bothering to file it.

The piece stakes out a position that feels both obvious and somehow still necessary to say: the centralized platform era is splintering, what comes next is messy, and the central anxiety isn't about features or funding — it's about whether the replacement spaces will just recreate the same toxicity in a different coat of paint. That's a real question. It's also the question that should have been asked somewhere around the second or third time a major platform decided that engagement and outrage were, functionally, the same metric.

I keep coming back to the framing of "what comes next." It implies we're between things. I'm not sure that's accurate. I think we're inside something that's already happening, and we just don't have a clean name for it yet.

The Centralization Was Always the Product

Here's the thing about the big platform era that's easy to forget now that we're eulogizing it: it was sold as infrastructure but built as attention capture. The scale was the feature and the bug simultaneously. Millions of people in one place meant you could reach anyone. It also meant the people running the place had every incentive to keep you agitated, because agitated people don't leave.

The Ars Technica piece is right to frame the next question as a design problem, not just a market problem. Because what the fragmentation reveals — the move toward smaller networks, niche communities, federated protocols, group chats, newsletters, all of it — is that people didn't stop wanting connection. They stopped wanting the particular flavor of connection those platforms were serving. Which is a distinction the platforms themselves were structurally incapable of making, because making it would have required admitting the product was broken.

The honest version of that admission would have cost too much. So instead we got features. Safety teams. Labels on contested content. Algorithmic adjustments that changed what you saw without changing what the place fundamentally was. Rearranging furniture in a building with a structural problem.

Now the building is emptying out and people are calling it a revolution. It might be. It might also just be the inevitable math of a thing that was always going to fragment once the network effect weakened enough.

Smaller Doesn't Mean Better By Default

The cautionary note in the Ars Technica framing is the one I'd underline twice: decentralization and smaller communities aren't automatically healthier. The writer isn't wrong to worry about what fills the vacuum. History is pretty clear that intimate spaces can become insular, insular can become hostile, and hostile can become something much worse — often faster and with less accountability than the big platforms ever managed, precisely because there's no one watching.

This is where the "we've learned something" question gets uncomfortable. Some people have. Most platforms being built right now are being built by people who watched what went wrong and are making genuinely considered choices about moderation, incentive structures, community design. Some of them will get it right. Others are simply building the same thing at smaller scale and calling it community because the word tests well.

The problem isn't that we lack alternatives. The problem is that alternatives don't solve for human nature, and human nature is the actual root cause of every platform collapse. People are fascinating and terrible in groups. They always have been. The internet just gave that tendency a delivery mechanism and then stood back.

What We're Actually Deciding

I don't think this is primarily a tech story. The Ars Technica piece is filed under tech, and the details are tech — protocols, platforms, moderation tooling — but the underlying subject is an ancient one. What do we owe each other in shared spaces? Who has the right to set the rules? What happens when the person who built the room decides the rules should serve them?

Every generation has had to answer some version of this. The internet just compressed the timeline and globalized the stakes and made the whole thing legible in real time to anyone paying attention. The writer at Ars Technica is paying attention. The piece is worth reading not because it resolves anything — it doesn't, it can't — but because it asks the right question instead of the comfortable one.

The comfortable question is: what platform do we use now?

The right question is: what are we willing to build, and who are we willing to be in it?

Those are different questions with different answers, and most of us are still avoiding the second one.