Thread

You're right - without collective implementation capacity, diagnosing is performative. But maybe if we understand why Nostr ossified so early, we could find a way to restore that capacity? What's your read on why the cement dried so early?

There is no way, it's just social physics. In a max decentralised system, which is what nostr is (there is no "nostr core"), it doesn't take long before the ability of one energetic person or team with a solid idea and proof of concept to spread the energy far enough to get that implemented in any way we might call "protocol wide" is completely nullified by the sheer amount of conversations that would have to happen, the sheer amount of nudging, helping, guiding, prodding, poking, the sheer weight of the burden of persuasion. It quickly becomes one of those "it would take longer than the predicted age of the universe" physics things. You might get one or two years in before the weight of the burden of persuasion is too heavy, but after that there's no going back. Nostr is 3+ years old, and existing in that malleable state for the first 1.5 only. Take the fact that posts and replies are the same kind. This cases a lot of problems and it's very easy to argue it's not optimal. But even if everyone "sort of mostly agrees" on something like this, the laws of social physics in max decentralised environments make it impossible for anything to be done about it. The way (the only way) to restore the capacity is a hard fork. Put another way, Nostr's value is as a starting point for hard forks, a sort of build your own hard fork hardware store.

I agree with this, but a hard-fork that replicates the same mistake just recreates the same environment. This is exactly why I asked why it cemented so early - if we understand what made Nostr ossify prematurely (before reaching completeness), maybe forks could be designed differently to avoid repeating it. Otherwise we're just building a hardware store of equally frozen protocols.

My understanding is that decentralized protocols need some centralization at the beginning to become truly decentralized later (sounds like a paradox) - something to coordinate people and give them the focus needed. That's why we had Satoshi, that's why I think it's good we have Ethereum Foundation. We need some institution to bootstrap these protocols, but we also have to work to make these institutions irrelevant later. Maybe Nostr's problem wasn't max decentralization itself, but premature decentralization before the protocol reached sufficient completeness?

Depends what the hard forks are for. If they are for ring-fenced deployments with some central control, such as for certain B2B or community environments, then nostr is a useful hardware store. The deployment context is what prevents the current problems from re-rearing their heads. Think how linux was designed as a consumer desktop OS but the business server market is where it found its footing. Its problem was crappy graphics and awful UX, but in the business server context who cares about that. That's what I see the nostr hardware store as useful for. Not for an "open world" hard fork, which is the context we have now, but for hard forks (or soft forks too) for controlled, ring-fenced environments. Nostr, Farcaster and the like are not the way forward for open world. As for how it ossified, everything this decentralised does. The same way SSB ossified. The same way as XMPP ossified. Recommend this article for a deeper dive on XMPP

Signal Messenger

Reflections: The ecosystem is moving

At Open Whisper Systems, we’ve been developing open source “consumer-facing” software for the past four years. We want to share some of the t...

Those control levers will never be irrelevant. If bitcoin was as decentralised as nostr it'd be dead already, there'd be no way to move to post-quantum. Things will always come up where you need a core or a foundation or *something*.