[Lightning-dev] Fee structure

ZmnSCPxj ZmnSCPxj at protonmail.com
Thu Mar 14 11:16:12 UTC 2019


Good morning John-John,


> > > That would not regard the size of the payments though? A payment leaving the balance at the far end would pay the same proportional fee as one close to the middle.
> >
> > "Proportional" means the fee paid is greater if the payment being forwarded is greater.
> > That is what "proportional" means.
>
> My initial point was that the fee would be proportionally higher for every satoshi used closer to the far end of the channel.
> So the fee would proportional to the state of the channel in addition to the size of the payment.

But the final state of the channel has a subtractive relationship from the payment size.
So I fear we have largely fallen into an argument on exact semantics and mathematical models that may not be particularly relevant to pragmatic use of LN.

The "standard" solution to balance concerns and manipulations of fee structure has, I think, been "just make the proportional feerate itself proportional to how much the other side owns on the channel".
Regardless of whatever mathematical model you use, in the long run this will achieve your goal of a balanced channel anyway, whatever inaccuracy it may have in the short term.

Again, JIT Routing is superior to this anyway, as you have direct evidence that rebalancing would be beneficial to you under the current network conditions.
Balance is not always beneficial: consider a balanced channel containing 0.75 mBTC on both sides, that cannot serve a forwarding request for 1.0 mBTC in either direction.
JIT Routing handles that case by simply transferring at least 0.25mBTC from some other channel, then serving the forwarding request.



Regards,
ZmnSCPxj



More information about the Lightning-dev mailing list