Author: Benjamin Mord 2018-04-20 20:43:20
Published on: 2018-04-20T20:43:20+00:00
In a conversation between Benjamin and ZmnSCPxj, they discuss the identification of a 'neighborhood' and establishment of an order within that neighborhood for the purpose of cycle formation. Bloom filters are used to define a neighborhood as a valuable contribution in maintaining useful competition and redundancy. March 23 proposal is preferred over March 24. The condition for 23 allows 3 members of a 5-member neighborhood to think they form a single 3-member neighborhood while the remaining 2 members think they are in a 5-member neighborhood that includes the other 3 members who have formed a 3-member neighborhood. Establishing a unique ordering for cycle establishment within a neighborhood is a much more ambitious undertaking, which becomes brittle when the network is dynamic. A less ambitious but more robust heuristic would be one where the probability of establishing a channel is proportional to the number of bits in common in the pubkey hash, normalized by the number of nodes currently observed. The algorithm already dynamically sizes neighborhoods to be small, with high probability of neighborhoods to be 3-5 members. The main benefit of the algorithm is that it flattens the number of channels a node will have and reduces centralization. If each node opens two channels to two different nodes, this algorithm will force all nodes to have exactly two incoming and two outgoing channels, while having similar reachability as random selection. Intermediate nodes already know two hops. Nodes could track the routes of payments that flow through their channels and could spot fees that seem both large and popular, and could use this information to identify under-served nodes to which a direct channel might be in order.
Updated on: 2023-05-24T22:22:09.767191+00:00