The Terminology used here is very confusing. There are also Signal the App, Signal the transport protocol, and both are not libsignal. libsignal implements the Signal Double Ratchet AKA Axolotl and other cryptographic protocols used by Signal.
A quick skim of loquaz lead me to the protocol description Nip 4 Direct Messages of nostr: https://github.com/nostr-protocol/nips/blob/743e43a8d4bf4a37022e3b6551524b12e7cc54a0/04.md . That not only exhibits the pitfalls explained in this blog post, it also doesn't even try to have the security goals typical of modern encrypted messengers. E.g. Nip 4 provides no forward secrecy, nostr overall does not yet have provisions for key rotation, nor an interaction flow for verifying identities.
It might be a good idea for nostr to adopt libsignal or vodozemac or something similar for encrypting messages, so that they can concentrate on getting the "more than" parts right and innovate where they can make a difference with their protocol.
Re: Nostr
A quick skim of loquaz lead me to the protocol description Nip 4 Direct Messages of nostr: https://github.com/nostr-protocol/nips/blob/743e43a8d4bf4a37022e3b6551524b12e7cc54a0/04.md . That not only exhibits the pitfalls explained in this blog post, it also doesn't even try to have the security goals typical of modern encrypted messengers. E.g. Nip 4 provides no forward secrecy, nostr overall does not yet have provisions for key rotation, nor an interaction flow for verifying identities.
It might be a good idea for nostr to adopt libsignal or vodozemac or something similar for encrypting messages, so that they can concentrate on getting the "more than" parts right and innovate where they can make a difference with their protocol.