I read something interesting on Scott Hanselman's blog the other day regarding WPF. It was in relation to activating and dis-activating some controls based on the value of other ones. And I "quote":

His proposed solution is taking advantage of WPF's extremely powerful binding capabilities. I "quote" again:

He's happy with that. I'm not! He's putting business logic inside the view. And it scares me to see things like this because it tends to non-testable and "obscure" problems.