Our first debate of 2017 will focus on a contentious hard fork proposal known as "emergent consensus," a series of changes to the Bitcoin protocol which introduce a node policy based heuristic that dynamically governs the maximum block size.

More specifically, emergent consensus (implemented in Bitcoin Unlimited) removes the concept of a fixed block ceiling and replaces it with a "complex hash rate tournament with variable semi-sticky frictions and path-dependent block acceptance*" by removing consensus rules and making them node/miner policy.

We will spend the first 20 minutes reviewing the current consensus discovery process in the Bitcoin protocol (Nakamoto Consensus) and compare it to the proposed changes in Bitcoin Unlimited. From there we will jump into the following questions to spark debate (feel free to post debate questions in the comments section below):

1) Should the maximum block size be a consensus enforced rule that all users must follow? If so, what purpose does it serve? If not, why?

2) Given one chain, two Bitcoin Unlimited nodes may validate it differently. What are the implications of this on node operation and the importance of out of band communications on consensus convergence? To what extent does Nakamoto Consensus rely on out of band comms?

3) Are non protocol experts able to actively monitor/modify their Acceptance Depth and Excessive Block parameters to best maintain consensus while also retaining the right to reject blocks? Can you just "set it and forget it?"

4) Are miners incentivized to avoid an AD/EB pairing that would cause them to orphan blocks/waste hash power? If so, does this conformity make them more susceptible to accepting blocks larger than they would otherwise?

5)If the miners have very different parameters how does user experience around confirmations change?

6) The complexities of adversarial behavior in emergent consensus is fairly difficult to reason about. Has it received enough peer review to go live?

7) Should Bitcoin Unlimited use the preexisting Bitcoin Improvement Proposal process?

--

*nullc