delvingbitcoin

BTC Lisp as an alternative to Script

BTC Lisp as an alternative to Script

Original Postby instagibbs

Posted on: March 25, 2024 13:30 UTC

The primary concern raised revolves around the challenge of utilizing Simplicity as a "Consensus Language," particularly focusing on its reviewability due to the substantial amount of code it entails.

The reference to the Elements1219 pull request, which shows a significant code change (+41238 lines added and -18 lines removed), highlights the complexity involved in implementing Simplicity. This complexity is not just limited to the sheer volume of code but extends to the expertise required to navigate and understand this ecosystem effectively. The suggestion posited is that rather than delving into the C/C++ code directly, it might be more beneficial to concentrate on understanding the formal specifications and the tools used for code generation.

Furthermore, there's an underlying suspicion that a major portion of the code difference observed, possibly over 80%, is attributed to components like COQ proofs and other supporting infrastructure. These elements are not necessarily integral to the consensus code itself but contribute significantly to the perceived complexity. The analogy of deleting mechanical proofs to simplify review processes, likened to cutting brake lines to prioritize speed, underscores the inherent risks and trade-offs involved in attempting to streamline reviewability at the cost of potentially compromising on necessary safeguards and thoroughness. This comparison serves to emphasize the importance of maintaining a balance between efficiency in review processes and the integrity and security of the code.