Remove OP_RETURN altogether? What is your definition of STANDARDNESS? To remove a default? If there is "no standard" than what is the default you propose? To refuse to run unless someone intentionally configures the parameter? All that could do is establish to a de-facto standard size.
It does nothing to address using outputs to hold data, which makes the utxo set demands ever worse. The OP_RETURN is almost a necessary feature to hash-stamp an out-of-band proof to a transaction. Yes, BIP110 is insufficient to stop spam, but it will implement resistance, and knots-style filters will add more resistance.
The only remaining alternative is to move to ZK proofs and/or encrypted UTXO destination addresses, not for privacy, but to eliminate the possibility of clear-text transactions and to provide liability protection through plausible deniability to node operators.
It does nothing to address using outputs to hold data, which makes the utxo set demands ever worse. The OP_RETURN is almost a necessary feature to hash-stamp an out-of-band proof to a transaction. Yes, BIP110 is insufficient to stop spam, but it will implement resistance, and knots-style filters will add more resistance.
The only remaining alternative is to move to ZK proofs and/or encrypted UTXO destination addresses, not for privacy, but to eliminate the possibility of clear-text transactions and to provide liability protection through plausible deniability to node operators.
1
