Author: Peter Todd 2014-03-22 08:47:02
Published on: 2014-03-22T08:47:02+00:00
Recently, there has been a discussion about proof-of-publication and the reduction of OP_RETURN length to 40 bytes before the 0.9 release. Additionally, it was noticed that OP_CHECKMULTISIG sigops were not taken into account, which could allow broadcasting unminable transactions and bloat mempools. The suggestion was made to ditch bare OP_CHECKMULTISIG outputs and switch to P2SH multisig scriptSig encoding, which would require Counterparty to change. It was also suggested that Bitcoin Core be forked to add anti-DoS measures and preferential peering for nodes with a broader idea of what is an "allowed" transaction. A graceful switchover would require applications to adopt format flexibility, and a good implementation would let PUSHDATA's used for encoding data be specified arbitrarily. Proof-of-publication is essential, and miners would be smart to support it as it is more secure to be embedded within Bitcoin than alongside it. There is a censorship risk to going the embedded route, but it would require explicit blacklists. Finally, MSC, XCP, and others should think for themselves before blindly trusting authority figures.
Updated on: 2023-06-08T15:32:58.883890+00:00