Author: jl2012 at xbt.hk 2015-09-03 07:57:09
Published on: 2015-09-03T07:57:09+00:00
The email thread discusses the implications of different threshold levels for a hardfork or softfork in relation to increasing or decreasing the block size limit for Bitcoin. The 75% rule is deemed meaningless as this is a pure relaxation of rules and there are no invalid Version 4 blocks. The implication threshold is unclear and it is suggested that the initiation of a hardfork should be based on its acceptance by the economic majority, not miner support. It is argued that a 95% threshold is unnecessary and an 80% threshold would suffice. Using a 20-percentile threshold to increase the block size is acceptable as it only increases when most miners agree. However, using a 20-percentile threshold to decrease the block size could be dangerous as it could incentivize a 51% attack on the minority who does not agree. Therefore, it is suggested that the thresholds for increase and decrease must be symmetrical. The use of a "hardfork bit" to signify the hardfork and/or combining the hardfork with a softfork is also discussed. A new proposal suggests replacing the static 1M block size hard limit with a floating limit that floats within the range of 1-32M. The initial value of the hard limit is 1M but can change by encoding a proposed value within a block's coinbase scriptSig. New hard limits are calculated at each difficulty adjustment period (2016 blocks) by examining the coinbase scriptSig votes of the previous 12,000 blocks and taking the 20th percentile and 80th percentile. The median of the resulting values is the new hard limit. Additionally, there are rules regarding version 4 blocks and the rejection of invalid blocks based on certain percentages.
Updated on: 2023-06-10T21:54:29.385248+00:00