What Gigi describes seems to me well oriented towards lessening technology usage, at least as how I get your points... as I see it, a nsec, that is light to handle, transportable and even disposable, has the advantage of lessening the burden of logins, username-password accounts, craft and storage, etc, let's say, there's a engagement decrease (or at least alternative kind of interaction) with the devices...
In this specific technology case. 1. I think a single "password/key" as access to your entire life is an absolutely terrible idea. 2. for the majority of people that means your entire life on someone else's servers (which, yes, is how things are now, and it's bad) I don't think you should have enough, or really anything to lose on someone else's servers or even on your own phone.
Technology is going to move forward with or without me, that's not lost on me, but I think there are simply some problems that should not be solved with technology, so I guess I'm arguing for caution?
1. Terrible because being a single point of failure? If so I agree. But also then, mitigation could be either to be cautious about what that key is securing, or managing a set of keys for specific purposes (although fair enough, we'd be back to square 1 in regards of device engagement 😅 )
With 2. I can't agree more as well. As for that matter nostr is also friendly towards running self hosting setups; that being said, self hosting could also be prone to single point of failure risks, ie, `centralizing' servers at home prone to $5 wrench's 🤔
I think since nostr just so many little components, I agree very self-hosting friendly. I've built my own signer and have a private relay (for testing) and so on. Self hosting is not cheap or easy, unless uptime isn't a big deal, so for sure. Anything that is too complicated for the average user to build/run risks centralization imo.
Oddbean new post about login | logout