On Sun, Jan 01, 2023 at 11:42:50PM +1100, Alfie John wrote: > On 31 Dec 2022, at 10:28 am, Peter Todd via bitcoin-dev wrote: > > > >> This way: > >> > >> 1. system cannot be played > >> 2. only in case of destructive halving: system waits for the recovery of network security > > > > The immediate danger we have with halvings is that in a competitive market, > > profit margins tend towards marginal costs - the cost to produce an additional > > unit of production - rather than total costs - the cost necessary to recover > > prior and future expenses. Since the halving is a sudden shock to the system, > > under the right conditions we could have a significant amount of hashing power > > just barely able to afford to hash prior to the halving, resulting in all that > > hashing power immediately having to shut down and fees increasing dramatically, > > and likely, chaotically. Your proposal does not address that problem as it can > > only measure difficulty prior to the halving point. > > > > ... Since the halving is a sudden shock to the system > > Is it though? Since everyone knows of the possible outcomes, wouldn't a possible halving be priced in? Re-read that I said. That explains why despite the halving being a forseeable event, there's no mechanism to "price it in" when it comes to hashing power. > > resulting in all that hashing power immediately having to shut down and fees increasing dramatically > > Which should cause that hashing power to come back because of this fee increases. Right now the total reward per transaction is $63, three orders of magnitude higher than typical fees. Sufficient fee increases to bring back hashing power in a scenario like that would cause enormous disruption to many things, including Lightning channels. -- https://petertodd.org 'peter'[:-1]@petertodd.org