Understanding Bitcoin Consensus Systems
One of the fundamental insights of Bitcoin is that information and consensus systems are inherently correlated. As a result, any knowledge or consensus in one direction can be translated into information or consensus in the opposite direction.
It is true that bitcoin users can possess their own information and consensus in either direction. But it is also true that if they try to do both at the same time, the information and consensus they may possess can be obsolete by the time they realize it. Given the speed at which information can be spread and the speed at which consensus can be dispersed, it is quite obvious that it is always possible to misinform or misperceive the content of an existing consensus.
To take an extreme example, imagine two individuals that both have a majority view about the fact that their neighborhood is unsafe. The first person may be correct about the danger. The second person is incorrect, but only because she failed to communicate the facts to her neighbors – i.e., she communicated them in a “safe” way.
It is impossible for this second person to know that the information she failed to share was erroneous. But suppose that her neighbor understood, and that she introduced a liveness check into her consensus system. Had the neighboring residents known that she had made a mistake, their assessment would have changed, and they would have been able to generate a liveness check that enabled them to overturn her vote. The good news is that their liveness check had a finite life cycle; as soon as the second individual realized that she had made a mistake, her neighbor’s liveness check would no longer be valid and would have to be replaced.
It is hard to assess how much bitcoin can be enhanced by ensuring that the majority of users have the same opinion as each other about consensus systems. If a communication platform contains a higher level of liveness, then bitcoin will be easier to exchange information. Otherwise, it will require bitcoin users to become more sophisticated in using their own versionsof consensus systems. The bottom line is that we cannot assume that the majority of users will have the same view on consensus systems.
Liveness in a communication platform is the same as liveness in a consensus system. Liveness means that when a user expresses an opinion, there is some amount of time before that opinion is confirmed or overturned. That amount of time is usually a relatively short period. It is reasonable to think that the amount of time could be reduced if all bitcoin users in a certain region of the world adopted a consensus system – however, that would lead to more isolation and much more of a challenge in preventing consensus among users.
With Liveness, the majority is in a consensus position much faster than it can be in isolation. And, once it has established a consensus, the Liveness comes to an end and the consensus can only be validated in its absence.
It appears that there are fundamental differences between the consensus systems that create consensus and the information systems that contain consensus. It is thus best to pay attention to two things: first, the nature of consensus in the first place; and second, the nature of consensus in a medium with a fast dispersion of information.