I like the idea of hosted channels. Solves low values payments. What are the regulatory implications for an LSP that custodies the sats until they end up in self-custody?
Discussion
There are exclusions for low amounts in some regs. There’s just little to no way to hold small, dust amounts of bitcoin yourself. Not a lawyer but there would be a clear upgrade path and threshold, there really isn’t an alternative 🤷🏻♂️
You could use ecash to do LSP fee credits. Before StashPay I thought about building a mint just for fee credits that LSPs could integrate. But I got stuck on the regulatory part. You’d have to prevent users from spending until they have enough sats to open a channel. And there is a loss of privacy, since you couldn’t keep track of amounts to stay within legal limits. Otherwise you’d just end up hosting a mint for folks and custody their funds without limits.
the need to be online is still the major issue to me after you upgrade to self custodied channels. not sure how the async receive stuff works.
Async receive takes that need away. The payment goes through from the pov of the sender and is held trustlessly at their lsp / wallet provider. When the receiver is available, they sign and receive it. No need to be online at the same time
pretty slick
Yeah, async payments would require broad BOLT12 support if I understand correctly. Basically the sender LSP would hold the payment until the receiving device came online.
seems like people are just going to move to cashu zaps before bolt12 zaps ever happen
Yeah cashu can iterate on the application layer more quickly. But choosing the initial mint is still an unsolved UX challenge. And mints get shut down constantly. I actually considered fedimint for stashpay, but chose the least worse option with Liquid (which is more likely to be there tomorrow).
I would trust liquid over random mints any day. still too much of a custody maxi to use liquid though lol
Lol, yeah. I see liquid as my pocket money where I can side-step onchain fee spikes like we had during the halving. I’m not religious about any particular tech though and will use whatever has the best trade-offs and UX. If we can solve zaps with self-custody tech, that would be ideal obv.
my biggest concern is losing the *option* for self-custody. when you build a protocol that excludes easy ways to do self-custody for people who want that, then that's a problem.
Well users can swap to onchain self-custody with integrated boltz. But I agree it’s not as seemless as hosted channels.
This was my original plan with the damus wallet. Occasionally swap to onchain when you hit a certain balance, but you lose the ability to spend those funds after you do that. Hosted + graduated channels are cool but require bolt12 for good ux and nothing is really built out yet 🥲
nostr:npub185h9z5yxn8uc7retm0n6gkm88358lejzparxms5kmy9epr236k2qcswrdp are we correct to assume that hosted channels would require async payments (and broad bolt12 adoption) to allow self-custodial zaps?
I mean hosted channels are just custodial wallets, they don’t *need* anything :). Still, I don’t think async payments are *that* required even for non-custodial - they solve an important problem but for apps people use, the on-notification hooks provide relatively good results, and the upgrade path to async means we can improve it as we go.
Yeah, if the receiver phone has an internet connection receiving on phoenix works relatively well while the app is in the background. I guess async solves for the cases where the phone is offline for extended periods of time. But since zaps are mostly not mission critical, but tipping for shitposts, it might work fine without async payments.
It would be a shame, nostr is built on owning your voice, your identity, shipping the money part off to shadow custodians so quickly doesn’t feel right
Yeah it’s sad, 98% of nostr devs seem ready to go full speed ahead in that direction though. i tried to build a non custodial vision but failed. Maybe i need to go back to working on lighting or other L2 tech again.