Floating-Point Nakamoto Consensus



Summary:

The author argues that delaying or preempting messages is cheaper and easier than producing a better fitness score. This creates a strange feature where an adversary can influence the size of both sides of the disagreement, creating a race-condition. The author's original statement that it is cheaper to introduce a large number of non-voting nodes than compete on mining power holds true, as long as they have a sufficient number of non-voting nodes. The argument against this point is that the attacker would have to split the network and have exactly equal mining hashpower on both sides, which is unlikely. However, the author argues that assuming all adversaries are crippled will not provide a good threat model, and both sides need to be more or less equal, but not exact. This metastable state can be maintained indefinitely, but there is no 100% assurance it will disappear. If a majority of messages are passed by malicious nodes, then Nakamoto's conjecture no longer holds.There is currently an active malicious-mining campaign being conducted against the Bitcoin network, where large mining pools delay the broadcast of a block in order to have a slight advantage on the formation of the next block. This withholding attack is already taking place because there is an economic incentive. The author suggests that although no proposed solution can prevent it completely, seeing that this bad thing would happen 1/2 as often is an absolute win.


Updated on: 2023-06-14T15:38:27.872012+00:00