I rarely come here. My posts don't have any reach, everything revolves around a small number of people, and it's very difficult to find someone to follow. This doesn't bother me, but it must bother those who expect centralized networks to behave in a similar way. In addition, the applications are unusual and the authentication systems created to solve the nsec/npub problem are confusing. If it's already complicated engaging a new centralized network, is even more difficult, let alone something so experimental.
Discussion
Another problem with the network: Because it is a protocol, everyone rushed to implement a zillion new things at the same time, without solving a single thing well.
If it's a protocol, you should focus on defining it and providing the tools to work with it. I don't see how it's productive to build a protocol from scratch while building high-level applications. That's insane, and that's why there are so many relays and clients that don't implement all the functionality. Nobody had to build the TCP/IP stack to create an email client.
"...the authentication systems created to solve the nsec/npub problem are confusing..."
this may be the biggest issue.
I'd like to see subkeys. Start a new account, have your identity create an attestation for it... that's it. If it gets pwned you can make an event disavowing that key and deleting the events created by it, then make a new one and move on
And if you wanted to, you could continuously rotate the key in order to limit the number of events that would be deleted. I have a PR open