There used to be this concept call a "usability study", where you got five or ten people who'd never used your product in a room and watched them use your produ

There used to be this concept call a "usability study", where you got five or ten people who'd never used your product in a room and watched them use your product. It didn't give the benefits of a panopticon, but it solved the problem without requiring a keystroke logger.