Good morning,

> I don't see why this wouldn't work as long as implementations on both
> ends supports channels multiplexing, like lnd or eclair do (didn't
> test it though).

Thank you, that is my understanding also.

> But being the accepting node, I wouldn't like receiving too many
> channel requests that never confirm.

I understand and this is a concern. Hopefully a later Lightning BOLT spec can 
provide some kind of cancel_channel message to indicate that the funder node is 
very sure that a particular funding transaction will never confirm.

> Also, most of the time I don't
> think the opening of a channel should be considered a time-sensitive
> operation (in general, but there are exceptions to this of course).

I think all blockchain operations should be considered potentially 
time-sensitive, including channel opening.

In theory it would be possible (but likely not easy) to develop a wallet which 
strives to keep transactions replaceable. Each blockchain send from the wallet 
would replace any unconfirmed transaction, and funding a channel would be 
similar to such a blockchain-level send (except the protocol needs to wait for 
funding_signed for all channel counterparties before broadcasting a replacement 
transaction).

I also think that a good part of the frustration around transaction fees at the 
blockchain level is at least partly due to the lack of ways to control 
transaction fees after a transaction has been broadcast.

Regards,
ZmnSCPxj
_______________________________________________
Lightning-dev mailing list
Lightning-dev@lists.linuxfoundation.org
https://lists.linuxfoundation.org/mailman/listinfo/lightning-dev

Reply via email to