CISA and Privacy

[1440000bytes]

@harding

The size is 355 vbytes, a 23% savings over the base case.

This will be below 5% if there is no change or more inputs are used.

@harding

The existing Bitcoin protocol heavily incentivizes batching. CISA makes batching only slightly more efficient, so I don’t think there’s any reason to consider it a privacy problem on that basis.

Batching is mainly done to pay multiple people in the same transaction irrespective of inputs used. The problem here is mostly related to inputs and not outputs. Multiple inputs in the same transaction that belong to us affect privacy in 90% of the cases.

Examples: Unnecessary input, Mixed types of inputs, Toxic change as input etc.

CISA demotivates coin control and incentivizes more inputs to be used in the same transaction. Coinjoin is less than 10% of bitcoin transactions.

@harding

Additionally, CISA reduces the cost of creating coinjoins and payjoins, two deployed protocols that improve privacy. It may also reduce the cost of other protocols that enhance privacy, including both current protocols (like LN channels closes with multiple in-flight HTLCs) and proposed protocols. The cost reduction is modest, but I think anything that gives privacy an advantage is worth considering.

Are you saying that users will do more coinjoin because of 5% discount but they wont do normal transactions in which all inputs belong to them?