I've already started noticing the kind of behaviour this book aims to eradicate that all of us developers are guilty of and I'm wondering what it will take to attenuate our high-held opinions of how software should work and prevent them leaking through into the (un)usability of software.
I can remember more than one occasion on which I've thought about the way a feature should work and concluded that it should be a true reflection of the underlying implementation. Talk about arse-backwards! The fatal flaw in this assumption was of course that the user's model of the underlying process will be the same as mine, if they have one at all! My personal model will have been completely skewed by my attempt to implement the feature: I will have been effectively forcing my own prejudices on the user.
So, for any work I've produced that causes users pain: I apologise and promise to try not to do it again. You're the king in this land and I'll do my level best to listen to you more in future.