Okay, I’m going to write my thoughts on Nostr, then nobody ask me about it ever again.
The guy who made it was on bitcoinhackers.org. He started building an ActivityPub backend in C. We were like “why C?” and he said “I want it to be portable and run anywhere”. He did a livestream coding it, where he was saying “don’t ask me why I’m building my own JSON parser. I’m not happy with the ones in C.” We were like WTF, use a library. Also, Pleroma is portable with OTP releases, there’s no reason not to use that. He acknowledged this and stopped the C implementation.
So unhappy with Pleroma, unhappy with the JSON parser in C, and unhappy with the ActivityPub spec he went on to create Nostr. At first I thought there’s no way this will catch on. But it turns out he’s a core contributor of Bitcoin, and was able to leverage his existing audience in cryptocurrency to popularize it. So now all the crypto bros on Twitter who never had a reason to care about the Fediverse care about Nostr. Forbes is writing about it, Elon Musk is acknowledging it, etc.
Is Nostr a good idea? 10 years ago, yes. It’s “simple” because we haven’t been building on it for 10 years. Between the two, I would have picked Nostr 10 years ago, but today it is a big distraction from accomplishing the goal of defeating big tech. It’s fragmentation and division, from a person who is incapable of (or doesn’t want to) work within an existing ecosystem.
So every time I see Nostr, I wonder if I need to drop everything I’m doing and make an ActivityPub server with decentralized IDs and client/replay paradigm to compete with Nostr. But I suspect even if I did, it would be ignored. It’s unfortunate he used his platform to weaken the Fediverse instead of make us all stronger.
Now there are calls for “a bridge between ActivityPub and Nostr” because it was an afterthought. It would have been so much less work, so much simpler, to just… make an ActivityPub client do the Nostr thing. So what’s my position on Nostr? No thanks.