Author: Anthony Towns 2022-03-11 04:46:45
Published on: 2022-03-11T04:46:45+00:00
In a bitcoin-dev thread, ZmnSCPxj discusses the differences between Lisp and Chia in terms of how they handle puzzles/solutions and cross-input signature aggregation. They note that the latter is more complex and involves more math and crypto than reversible compression of puzzles/solutions. However, they also state that cross-transaction relationships are more important than cross-input ones in Bitcoin's context. The conversation moves on to discussing jets, which are costed differently at the consensus level, and how generating Lisp code with Lisp code does not receive special treatment in comparison to using jets. PTLCs are suggested as being more space-efficient than HTLCs, but they are not ideal for Chia due to a lack of secp256k1 operations and an expensive extra ECC operation. In terms of covenant complexity, ZmnSCPxj expresses concerns about bugs sneaking in and how it is always better to have a bug at the wallet level rather than in a new sighash feature that has been soft-forked into consensus. They also state that a covenant language implementation may not necessarily be complicated and could make implementing ANYPREVOUT/CTV/TLUV/EVICT/etc simpler than doing so directly in C++.
Updated on: 2023-05-22T17:50:09.581910+00:00