Relax OP_RETURN standardness restrictions

The discussion revolves around the technical aspects of implementing a soft fork in the Bitcoin network, specifically addressing the potential for decreasing the maximum block weight from the current 4 million to 1 million and adjusting the witness scale factor from 4 to 1. This approach essentially reverses the block size increase that came with the introduction of Segregated Witness (SegWit). For nodes that have not upgraded to SegWit, this change would manifest as a reduction in block size since these nodes are unable to perceive witness data. The proposal acknowledges the absence of a reverse scenario where such adjustments could apply, suggesting an exploration through modifying Bitcoin Core's tests to affirm this uniqueness.

A significant concern highlighted is the confiscatory nature of this proposed soft fork. Transactions that were previously valid, such as a pre-signed transaction exceeding the new 1 MB limit, would no longer fit within a block post-implementation. This raises practical implications for users holding such transactions, necessitating either the assistance of a miner or the use of an accelerator to process them due to their non-standard status under the new rules. The mention of historical criticism towards soft fork proposals, including those perceived as less likely to impose confiscatory effects, underscores the sensitivity and complexity surrounding consensus changes within the Bitcoin ecosystem.

Previous discussions and proposals related to reducing the block size, while maintaining the SegWit discount, are also referenced. One such proposal, albeit never formalized through a Bitcoin Improvement Proposal (BIP) number and lacking broad support, aimed to introduce a byte-based ceiling without traction within the community. This context suggests a skepticism towards the feasibility and desirability of pursuing block size decreases as a method for removing the witness discount, given the challenges in garnering necessary consensus and the interdependent relationship between block size and witness data handling.

For further reading, the original discussion can be explored at Bitcoin Stack Exchange.