19:21:42 I hear you. I'm (obviously) aware that the number of inputs is not a wallet-specific parameter, but a function of the enote set and the value of the transaction. I don't think the privacy gain of mandating at least 2 inputs can be easily fit into your framework, because it's not a non-standard wallet behavior. I guess if you approached it counterfactually, i.e. suppose that 2-in t 19:21:42 xs are the standard, and 1-in txs are privacy defects, you could quantize it. 19:26:21 *suppose that 2-or-more-in txs 19:27:10 yes, that casts a doubt on the worthiness of requiring at least two inputs