Author: Billy Tetrud 2022-02-27 16:59:54
Published on: 2022-02-27T16:59:54+00:00
Paul argues that largeblock sidechains should be reconsidered as they can increase blocksize for those following the sidechain. He also suggests that users can pay more or less for decentralization based on their preferences. However, decentralization is either enough to stop a class of attack or it is not, and increasing blocksize ten times larger is a poor design choice. Even if individuals only put a few satoshis in such a sidechain, stealing them would still have a significant impact on society. Instead, Paul proposes learning from past mistakes and creating new sidechains from scratch using improved design principles.In response to Paul's argument, ZmnSCPxj discusses the technicalities of changing the ownership set in sidechains and how it requires a 32-byte commitment in the coinbase. He explains that if mainchain miners are separate from sidechain validators, then an additional `OP_BRIBE` transaction is needed. ZmnSCPxj also points out that continuous operation of sidechains and ordinary payments require one commitment of 32 bytes per mainchain block, whereas continuous operation of channel factories has zero bytes per block being published. He suggests that optimizing for continuous operation seems like a better tradeoff.ZmnSCPxj goes on to explain the differences between merge-mined sidechains and LN-with-channel-factories, highlighting that with channel factories, payments have no onchain impact at all. In contrast, with merge-mined sidechains, all activity requires a commitment on the mainchain. Additionally, starting a new sidechain with Drivechains requires implicit permission from miners, while new channels and channel factories in LN do not require any permission.
Updated on: 2023-06-15T16:38:13.618698+00:00