From o.thyfronitis at ed.ac.uk Fri Nov 22 11:13:36 2019 From: o.thyfronitis at ed.ac.uk (Orfeas Stefanos Thyfronitis Litos) Date: Fri, 22 Nov 2019 11:13:36 +0000 Subject: [Lightning-dev] A proposal for up-front payments. In-Reply-To: References: <87ftj33w2z.fsf@rustcorp.com.au> <87d0e5zwt8.fsf@rustcorp.com.au> <20191106155733.e2b2ttxs4wsfoqpg@erisian.com.au> <87sgn0ux9w.fsf@rustcorp.com.au> <20191107113735.72mxwsrcaidydflk@erisian.com.au> <875zjvum7n.fsf@rustcorp.com.au> <20191108123236.kmi4dyqyix2hztyo@erisian.com.au> <87pnhzqpg3.fsf@rustcorp.com.au> Message-ID: <65c97be9-7d84-565d-e8d7-da82869a80b5@ed.ac.uk> Good morning ZmnSCPxj, > requiring a fee is equivalent to requiring proof-of-work, incentive-wise. Not necessarily, given that 1) there is a finite bitcoin supply but an eventually infinite PoW supply (relevant in the unlikely case fees are burned) 2) sats are transferrable, whereas PoW isn't (relevant in the case fees are paid) On the other hand, there exists this paper with the fancy name that claims using PoW for spam prevention in the context of email (the original context in which PoW was discovered) is ineffective due to the high per-message PoW required to beat spam [0]. Therefore we have to see whether this paper applies to LN as well before going down that road. Spam prevention in an unauthenticated system is much more complex than it seems at first, because it boils down to avoiding the Sybil attack, (one of) the most difficult problem(s) in such systems. A (traditional) reputation system in essence enables authentication (eww), per-message PoW might be too expensive, and per-message fees seem to have incentives issues and are kind of misaligned with LN's aims. Maybe I've missed something, but what makes spam in LN a bigger problem than it is in every other p2p network out there? Why won't traditional bad activity thresholds do the job? I don't think spam is something that will be completely wiped out, only contained. LN should provide for many orthogonal spam prevention measures (local tunable activity thresholds, gossipable reputation systems (eww), per-message fees, per-message PoW) with sensible defaults to allow users to experiment and choose what is best for them, but that may lead to unacceptable protocol and UI complexity. What a tradeoff... Best, Orfeas [0] https://www.cl.cam.ac.uk/~rnc1/proofwork.pdf -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.