Author: Christian Decker 2015-05-13 19:14:57
Published on: 2015-05-13T19:14:57+00:00
In an email exchange between Pieter Wuille and Christian Decker on May 13, 2015, they discussed the proposals for addressing transaction malleability. While Decker argued that using legacy references to transactions had utility, Wuille contended that any complex interactions with multiple unconfirmed transactions waiting on each other should rely solely on normalized transaction IDs to avoid suffering losses due to malleability. Wuille described the SIGHASH proposal as adding a new CHECKSIG operator to script, which removes the scriptSigs from transactions before hashing and replaces the txids in txins by their ntxid, but Decker expressed skepticism about its hackish nature, the fact that it added more data to the script, and that it could invalidate a block if relayers changed the txids. Wuille acknowledged that the normalized transaction ID required a hard fork and the whole world to change their software, which would incur large up-front deployment costs, but believed that its advantages would prevail in the long-term. He also suggested increasing the transaction version, so every transaction would be either in one index or the other, reducing the deployment cost to almost nothing.
Updated on: 2023-06-09T20:59:55.603693+00:00