The "Breathalyzer Test"

by Ed Yourdon

Sooner or later, every IT organization experiences the unpleasant phenomenon of the "doomed" project. For obvious reasons, it's better to find out about these sooner rather than later, because it's better to kill the project before too many careers have been ruined and too much money has been wasted. Unfortunately, it's not always easy to tell whether the project has developed a fatal case of gangrene, and organizational politics often prevent a frank and open discussion of the situation. The project manager, for example, is often the last person to admit defeat; he or she is likely to continue arguing that with a little more time, a few more people, and a little bit of luck, the project can be brought to a successful resolution. The project sponsors (or end-users) may have invested so much money that they don't want to give up, and various other high-level managers may also be reluctant to face the unpleasant prospect of coping with a project failure.