• The Pattern Nobody Wants to Name
• What Mastodon Got Wrong
• What Bluesky Got Wrong
• What Nostr Does Differently — At the Protocol Level
• The Third Ingredient: Bitcoin
• Why This Time Is Different

Every few years, a wave of frustration rolls through the internet. Users get burned by a platform ban, a policy flip, an acquisition, or a CEO who types “WAFFLES” in response to a user safety concern. They flood to the nearest Twitter alternative. They post. They onboard their friends. Then, quietly, they drift back.

It happened with Mastodon in 2022. It happened with Bluesky in 2024. It will happen again , unless something structurally different enters the picture.

That something is Nostr.

§

§The Pattern Nobody Wants to Name

Mastodon peaked at over 2.6 million monthly active users in November 2022, riding the first major Musk-era exodus from Twitter. By early 2025, that number had collapsed to under 690,000. The users didn’t delete their accounts out of anger. They just stopped showing up.

Bluesky had a louder second act. After the 2024 U.S. election, it shot to the top of the App Store. Celebrities joined. Journalists migrated. It felt, briefly, like the future. Then by the end of 2025, active posting dropped by roughly 40%. The growth flatlined.

Two platforms. Two surges. Two slow deflations.

The question isn’t why people left X. That part is obvious. The question is why they kept coming back.

§

§What Mastodon Got Wrong

Mastodon is a principled project. Genuinely decentralized, non-commercial, community-governed. On paper, it checks every box.

In practice, it asked too much of ordinary people. To join Mastodon, you had to choose an “instance” — a server operated by a stranger, with its own rules, its own community norms, and its own risk of shutting down one day. Your identity lived on that instance. If it disappeared, so did you.

That’s not decentralization for users. That’s decentralization at users. The cognitive overhead was fatal. Most people don’t want to understand infrastructure. They want to speak, be heard, and go on with their lives.

Mastodon solved the censorship problem but created a complexity problem. And complexity, at scale, is just a slower kind of failure.

§

§What Bluesky Got Wrong

Bluesky understood the complexity problem. It built something clean, fast, and familiar. And for a moment, it worked.

But Bluesky made a different kind of mistake, one that’s harder to forgive because it was more deliberate. It presented itself as decentralized while remaining, functionally, centralized. The AT Protocol is real. The federation promises are real. But the identity layer, the moderation layer, the infrastructure ,all of it still runs through Bluesky the company, in Bluesky’s servers, under Bluesky’s terms.

That’s not a technical footnote. It’s the whole problem. When the company’s moderation broke down , banning accounts selectively, dismissing user safety concerns, letting harassment circulate while silencing critics , there was no escape hatch. Users couldn’t fork the moderation. They couldn’t move their identity. They could only leave, and lose everything they built.

A platform that promises decentralization but delivers it conditionally isn’t a solution. It’s a delay.

§

§What Nostr Does Differently — At the Protocol Level

Here is the structural fact that makes Nostr different from everything that came before it:

Your identity is a cryptographic key pair. It belongs to you. It cannot be taken away.

There is no Nostr Inc. There is no Nostr server you must trust. There is no terms of service that can suspend your account. You sign your content with your private key. Anyone who wants to verify it can. Any relay can carry it. Any client can display it.

This is not a feature. It is the architecture.

When Mastodon bans you, you lose your followers and your history. When Bluesky bans you, same result. When a relay on Nostr blocks you, you find another relay. Your keys, your content, and your social graph remain intact , because they were never held by anyone else.

§

§The Third Ingredient: Bitcoin

Mastodon and Bluesky both treated the monetization question as someone else’s problem. Nostr didn’t.

Through the Lightning Network, Nostr clients like YakiHonne enable direct value transfer between readers and creators, no platform taking a cut, no subscription paywall, no advertiser in the middle. A reader sends sats to a writer. The transaction settles in seconds. No bank. No Stripe. No terms of service review.

This changes the entire incentive structure of social media. Platforms like Mastodon and Bluesky still rely on the attention economy , engagement metrics, content that provokes, algorithms that reward outrage because outrage retains users. Nostr doesn’t need that model. When value flows directly from reader to creator, the incentive is to write well, not to write angrily.

§

§Why This Time Is Different

The failure of Mastodon and Bluesky wasn’t a failure of values. Both projects genuinely wanted to build something better. The failure was architectural. They tried to fix social media by changing who runs the server. Nostr eliminates the question of who runs the server.

You cannot “enshittify” a protocol. You cannot acquire a key pair. You cannot ban a cryptographic identity.

Every platform that has ever failed users from early MySpace to late Twitter failed for the same reason: someone, somewhere, had the power to change the rules. Nostr is the first serious attempt to make that power structurally impossible.

The wave always returns. But this time, there is somewhere worth landing.

Published on YakiHonne — the decentralized social client built on Nostr & Bitcoin.