Author: Christopher Jamthagen 2015-07-30 10:01:17
Published on: 2015-07-30T10:01:17+00:00
In this email conversation, two individuals - Rusty Russell and Christopher Jamthagen discuss the Lightning Network protocol. They discuss a scenario where Alice wants to pay Carol some amount through Bob. Carol gives Alice H(R), which Alice then uses to update her commitment transaction with Bob. Bob also updates his commitment transaction with Carol. However, Carol withholds R, forcing Bob to broadcast his commitment transaction with Carol. This puts a delay on Bob's ability to spend the HTLC output, but Alice can spend the HTLC output if the CLTV has expired. The conversation then moves on to discuss Gregory Maxwell's timestop feature, which delays the expiration of CSV in the case of full (or near-full) blocks. If this is used, the counterparty can fill the blocks for a limited amount of time until their CLTV has expired, and then take the HTLC output. The individuals discuss whether it is desirable or possible to implement the timestop feature for CLTV as well. They also discuss the CSV delay amount, which controls the worst-case HTLC that can be accepted. They suggest that a CSV of a few hours should be okay if one is online all the time. Additionally, they discuss the possibility of outsourcing blockchain checking to miners to keep an extra eye on the blockchain for cheaters in the Lightning Network.
Updated on: 2023-05-23T18:58:41.904604+00:00