Author: Luke Dashjr 2016-12-10 21:41:57
Published on: 2016-12-10T21:41:57+00:00
On December 10, 2016, a discussion took place in the bitcoin-dev mailing list regarding the new merkle root algorithm described in MMHF BIP. Tier Nolan suggested that any new merkle algorithm should use a sum tree for partial validation and fraud proofs. However, he also mentioned that adding new trees means more hashing to validate blocks, so it would be better to keep it at a minimum. Johnson Lau had earlier stated that something not yet done was the new merkle root algorithm. He wanted to use the algorithm described in MMHF BIP. Luke Jr. responded by asking if there was something special about 216 bits. The 448 bits total target is to optimize the tree-building. He further suggested that it would be even better to make the protocol extendable, allowing blocks to indicate new trees and legacy nodes ignoring the extra ones. In this way, the segregated witness tree could have been added as an easier soft fork. Finally, he added that the sum-tree could be added later as an extra tree, but it is important to keep the hashing to a minimum.
Updated on: 2023-06-11T20:46:06.096017+00:00