With the L2s there seems to be this dilemma between trust and cost. Non-custodial Lightning sits at one extreme where it's basically trustless but incurs high cost (channel management, routing). Something like Fedi sits at another extreme where trust is higher (federated, hard to audit) but payments are cheap and frictionless. This trade-off is not a bad thing, it's just reality. The ultimate form of sovereignty in Bitcoin is UTXO ownership, but this is necessarily expensive and most people will not be able to afford it.

I think that the non-custodial Lightning UX is skewed towards favoring the sender. All you really have to do is fund a wallet on Phoenix or whatever and hit send. Receivers on the other hand have a much harder time because they require in-bound liquidity, need to be online, etc., which of course is why custodial wallets are so popular on Nostr. To me Fedi looks like a strict upgrade from custodial Lightning because it has the same advantages and more (such as no routing required). Much easier to use on Nostr.

I suspect a handful of banks will emerge for liquidity and trust reasons, similar to how there are only a couple of dominant stable coins and ETH LSDs. Not really sure why we'd end up with thousands of community banks, as a merchant would have to vet all of them to be confident that their e-cash is backed by BTC. Perhaps there's something I'm misunderstanding here though.