Replies (15)

Ok, that's a point then. But with on-chain subs, it still could get dangerous because of chainalytics. Bitcoin's blockchain is much more monitored than Nostr. At least, that's what I think.
Technically, it is harmful. 1) people receive sats they don’t even know they received. 2) nsecs are not that secure, all of them are hot, and we are just adding more layers of risk on a single string of text. 3) it’s very bad for privacy. 4) it will create a large amount of utxo dust Lightning can handle 100K sats. So not really helpful to do this.
How do you think people should use different nsecs then? Of course, there can be multiple nsecs for different purposes. But using a different nsec for each post? This improves privacy, and I have made apps myself that do this when not logged in with an existing profile. But without having a seed for this, saving them all manually is nearly impossible.
So purpose-based, right? I still could use my profile here for sharing things publicly? But... the same way you publish a Lightning address to your kind 0 profile data, why not also put in a Bitcoin address field? This could support single-key addresses (like P2TR, P2PKH etc., still bad for privacy, but not tied to your nsec), but also PM8T (BIP-47) or silent payment addresses. You want ro see these as well, right?
One thing I think worth pointing out is that zaps on lightning are all public; sure. But when you then send those sats somewhere else, still on lightning, there's no continued visible trail. This is not the case on chain. It then becomes something the user is either proactively taking steps to break the trail on, with something like a coinjoin, or swapping it into lightning, or more likely, given that we'd be looking at these for the LEAST savvy users who don't have lightning set up, just left there to track absolutely everything about that UTXO. I do totally get coding things just to see if you can. And it's sort of cool that you can. I just think this is one of those things that, upon realizing it can be done, probably still shouldn't be.