I agree that prototyping the front end is good starting point and a great way to build end-user confidence. I think what the article is lampooning is a project

I agree that prototyping the front end is good starting point and a great way to build end-user confidence. I think what the article is lampooning is a project that develops the user interface to the exclusion of everything else. Many specification flaws only come to light when developing the back end -- which may, in turn, require changes to the already-complete front end, wasting time. In general, the article makes fun of methodologies by picking a handful of software design approaches and pushing them to absurd extremes, hoping that the rest just falls into place. It's slightly amusing (and entirely consistent with other methodologies) that some of FAD's approaches may actually be helpful when applied rationally in appropriate situations.