He has a competing project that works in theory but not in practiceā¦
So he comes up with ideas as to why a project that works in practice shouldnāt work in theory.
Discussion
Iāve tried hard to really understand the essence of the Pubky projects but I canāt see any fundamental difference with nostr. For example, storing resource records in the Mainline DHT, the same can be done with signed events. As for pubky domain names, itās easy to integrate npubs into the existing DNS system
Youāre probably right that itās functionally the same with less openness and centralization.
Imagine you successfully pitch your āopen social networkā to Tether Ventures, get funded, and then see this wildly simple open protocol just start to take off and eat your lunch with no CEO. I suspect thatās the level of cope weāre dealing with here š .
*more centralization
(but i assume you caught my drift)
They are all just keys right?
What makes a nostr key special? Well, it uses the Bitcoin curve.
What makes a pubky key special? Well, it uses the ed25519 curve.
Nostr's curve allows some sort of handwavy Bitcoinyness.
Pubky's curve allows interoperation with Mainline DHT and other existing relevant crypto.
The result is Nostr always relying on censorable DNS to find anything, making keys disposable and hot. Where Pubky can actually use keys as persistent entities, and keep them cold.
Keys can be search on any network, but not every network can be trusted.
(This ONLY covers the differences for the identity layer, we could continue with how data is transmitted and aggregated, and then how search and coordination differ ... but i wont bother because i know you dont actually care)
I donāt follow the argument on how ed25119 helps keeps the keys cold.