Defensive Consensus: Getting to a Multi-Implementation Bitcoin Network
Jason Dreyzehner

Drawing from what works in other problem domains is the most logical path if you believe that information assurance is important in Bitcoin. Those that think that Bitcoin (the protocol, not the unit of account) should have a homogeneous implementation in order to minimize risk to the ecosystem, must also be against any further changes to that homogeneous implementation. The problem is, the very moment the hash of the binary changes because a bug is repaired or a new feature is added, you now have heterogeneity. At this point, you need something like defensive consensus. Further, making allowances for limited heterogeneity so that one single code base remains the “reference client” severely choices in a wide range of this problem space. The argument against these thoughts always seems to be “Well Bitcoin is an open-source project. Anyone can contribute code; the process is very egalitarian.” To which, I would reply, “Clearly you have 1) never participated in the process in which you speak or 2) are disingenuous.” The fact is, pragmatically, there is only a select set of people that can actually merge code into the master branch and there are a small set of people that can publish code on They are the invisible filter that certainly employ their editorial control. Whether this power is used for “good or evil” is certainly NOT the point. The mere fact that any person has this sort of control is THE PROBLEM. This is a bug and not a feature. Allowing a “priesthood” of gate-keepers who consider themselves ordained ministers of the implementation of Bitcoin smacks of Government. In the immortal words of Ronald Reagan, “Government is not the solution to our problem, Government IS the problem”. Great article!

Show your support

Clapping shows how much you appreciated Chris Kleeschulte’s story.