Author: micaroni at gmail.com 2022-04-28 08:03:53
Published on: 2022-04-28T08:03:53+00:00
The bitcoin-dev mailing list is currently discussing the issue of soft fork activation and how to measure user support for proposed soft-fork changes. Currently, all forms of activation involve miner signaling in one form or another, which has frustrated some users. One proposal is to allow transactions themselves to signal for upgrade by mapping free bits in the transaction version field to the signaling bits in the block header.However, there are concerns about potential manipulation and false consensus. Some members of the mailing list believe that the consensus should follow the current line of discussions and tests carried out by experts, with the most important developers having the most weight in discussions. They argue that consensus simply means there are not at least two or three important people opposing the idea with solid arguments.Others worry that this approach gives too much control to developers and doesn't adequately represent user preferences. There are also questions surrounding the issue of soft fork activation, including what the threshold for saying "this consensus change is ready for activation" should be, whether it changes based on the nature of the change, and if different constituencies have a desired minimum or maximum representation in this "threshold. "Additionally, tests must be devised to measure levels of support directly, and there is concern that these tests can be gamed. Ultimately, the mailing list members agree that technical consensus alone is not enough to assess user consensus, and more rigorous methods are needed to accurately represent user preferences without giving too much power to any one group or constituency.One proposal suggests a transaction signaling system that allows users to have sybil-resistant influence over miner decisions. Transactions can be included in a block regardless of the block's signaling vector, but a transaction signaling in the negative must not be included in a block that signals in the affirmative. This mechanism is measurable completely en-protocol and doesn't require trust in institutions that fork futures would.The proposal has several potential tweaks such as including a notion of negative signaling and making it such that miner signaling must be congruent with X% of transactions. Some anticipated objections include whether this proposal affords us a better view into consensus than we have today, whether it can be gamed to give us a worse view into consensus, and whether it measures the right thing.The proposal requires its own soft fork, but before considering how to make this happen, the forum should discuss whether or not it's a good idea.
Updated on: 2023-06-15T19:49:40.704269+00:00