pull down to refresh

There's been solid progress with these main approaches in the past year, hopefully the same ones you referred as "suggested ways":

Trampoline / LSP-mediated approach where an intermediate node holds the HTLC until the recipient comes back online. Éclair has had implementation work here for a while.

PTLC-based async payments that is the fully trustless approach favored by Matt Corallo. It uses Point Time Locked Contracts so the holding node can't steal funds.

Onion message signaling where the sender's LSP holds the payment and waits for an onion message from the recipient's LSP indicating they're back online to release the payment. This is the approach LDK has been building toward and Bitcoin Optech has a great topic page tracking all of this here https://bitcoinops.org/en/topics/async-payments.

Voltage also has a good overview here: #295133, and a more recent post on why this matters for mobile/self-custody nodes #1287485.

A broader look here by @roy and team #399604

Here few more:

But do not exclude:

"There's no such thing and never will be, anything above the chain is inherently interactive. These are immutable physics."
@justin_shocknet #1287707

... that somehow is right, we can perhaps just simulate it, giving the user the illusion "Async" or "Offline" payments in Lightning. Good UX~design can do that.

I have heard about PTLC, but I didn't know it can be used for async payments. However as far as I know there is no PTLC implementation yet. Or is it?
The onion message signaling between LSPs will work good enough, but I haven't heard it mentioned in two years. I think it can work well enough, provided nothing better is available.
The first link from voltage is dead. The second one does give hope that the LSP way of async payments may be available soon. I hope Zeus implement it since this is the wallet I am using (although I find it feature rich but very buggy).

reply

Thanks for the detailed breakdown, @supratic. Appreciate you linking those resources.

The trade-off between trustlessness (PTLC) and UX (LSP-mediated) is indeed the core challenge. Simulation of "offline" payments might be the pragmatic bridge we need until protocol-level support matures.

Great to see the discussion kept alive. Will dive deeper into the Optech links you shared. 🫡

reply
we can perhaps just simulate it

That's called trust, its not a UX challenge... just a narrative violation.

When people offer solutions they're not trying to improve UX, they're trying to trick the user (or themselves) because they don't want to confront a basic truth.

reply