Giving teeth to expected EC disabling: P2XX(-T)(-ML)

In a recent discussion on the Bitcoin Development Mailing List, there was an exploration of the potential errors in assumptions about the robustness and security implications of future cryptographic methods. The main point of contention centers around the comparative ease with which someone could influence the hash output using SHA2 versus how a canary-solver would operate. It was initially suggested that poisoning the future hash function (H=SHA2(G)) might be simpler than previously thought, contrary to the more complex tasks facing a canary-solver who, it turns out, would require the same computational effort as classical collision finding, specifically 2^128 work.

Further insights were offered regarding the use of unbiased data sources such as the hash of the genesis block or the first block where the canary mechanism is implemented. These suggestions were credited to Tadge Dryja, whose presentation can be viewed here. The conversation also touched upon the feasibility of moving to a 192-bit system and its practical implications. Such a transition could potentially extend the useful lifespan of certain cryptographic practices.

Additionally, there was a related thread focusing on the implementation of distributed key generation and Zero-Knowledge Proofs (ZKP) for the canary system. However, this method was recognized as less trustless compared to Non-Uniform Memory Access (NUMS), raising questions about its viability or the possibility of alternative solutions not yet considered. This part of the discussion emphasizes the ongoing search for more secure and efficient cryptographic techniques within the community.