New job opening at Bluesky (part-time; multiple positions available).
“Looking for useful idiots to legitimise us so we can achieve network effects before we screw over our user base.
Right candidate can state ‘I’m willing to give them the benefit of the doubt’ in public with a straight face in full knowledge that we’re a Silicon Valley outfit backed by the founder of Twitter.
The thing I like about #Nostr compared to #ActivityPub is it's secure by default. No need to put public keys in your bio and hope your admin doesn't MITM your conversations.
I like how flexible/extendable #ActivityPub is compared to #Nostr. You can just make new types so long as you have a URI to describe them. Whereas in Nostr they have to be approved by fiatjaf etc and they probably won't be approved if they "shitcoin". IIRC.
In case you've seen #Nostr accounts – the decentralised social protocol where coiners recently moved over to – being bridged into ActivityPub fedi: One popular bridge mostr.pub is – who could have guessed – developed by Alex Gleason, part-time egdelord, main dev of soapbox, fired from Pleroma, co-founder of spinster and enjoyer of supposed swastika “shitposting”.
If you decide to use that software, just be aware whom you might need to interact with for bugreports.
Recently, #Nostr protocol has started to gain traction, apparently because a former Twitter CEO started promoting it.
In this post I want to compare it with #ActivityPub because some people think Nostr is superior.
Advantages
- Censhorship resistance. If one relay prevents you from communicating with your friend, you can easily add another one. - Identity is not attached to a particular server.
While these things are good and important, there's nothing in ActivityPub spec that prevents developers from implementing them. Nostr proponents are lying about that, even on their official website. Chatternet project clearly demonstrates that you can use DIDs with ActivityPub. It can't connect to existing Fediverse platforms but I'm implementing similar solutions in Mitra in a way that preserves backwards compatibility as much as possible.
Disadvantages
I won't talk about the problems that can be solved, because obviously Nostr ecosystem is very young. But in my opinion it has three fundamental flaws:
- Clients can hide important protocol details from the user. Choosing relays and managing private keys is burdensome, so clients that don't force user to think about this will be more popular. As a result, people flock to sites like iris.to and this will lead to extreme centralization. This happened to all other web3 social platforms, and we can already see this happening to Nostr. - Nostr was created by bitcoin cultists and they have complete control over the NIP process. If you want to use any other currency there's always a risk that your proposals will be censored. And tying entire ecosystem to a single non-fungible coin is generally a bad idea. - Nostr is not based on web standards. This means it won't benefit from the adoption of emerging standards like DIDs and verifiable credentials. Reinvention of the wheel leads to unnecessary fragmentation and harms everyone.
So I see no reason to build anything on Nostr. Also, despite the hype I doubt that it will ever become more popular than Fediverse because ordinary people and organizations don't care about "censorship resistance" and "zaps". For them Fediverse is good enough.