SHAPE is an acronym for Software as a Human Activity Performed Effectively and is also the name of a web based discussion community devoted to issues in project management. The participants in the discussion are some of the leading figures in the area of the management of software projects and this book was constructed by selecting some of the more profound points made in the online debate.
What is most interesting about the discussion is that it deals with management situations rather than being restricted to software projects. The point I found the most useful is the description of serious failures that have occurred. Generally, when the problem begins, the decision makers are receiving accurate data that clearly indicates that a failure is imminent. However, it continues to progress and become critical because those receiving the data find it difficult or impossible to believe the data until it is too late. This is a very common occurrence in the software development world, as often everyone from the senior managers on down choose to ignore the warning signs that the project is moving towards failure. Even worse, anyone who breaks ranks to raise the issue is censured or even terminated. Finding a solution to this category of problem is probably the most difficult of all managerial problems to solve. Such a complex problem is not easily resolved, but the advice here will certainly help.
One other discussion that was of great interest is the one about the sinking of the Titanic. In fact, I learned some aspects of that most catastrophic of failures from the SHAPE discussion that I was not aware of, although some of the discussion is a bit unusual. It turns out that the limited lifeboat capacity was due to a redefinition of their purpose. Since the ship was unsinkable, the only possible use for the boats was to ferry passengers off in the event the engines were to quit. The most unusual point in the entire book was a dialog thread where the debate point was whether the attempt to avoid the iceberg was a mistake. It is argued that it would have been better to have rammed the iceberg, which would have severely damaged the ship, but not enough for it to sink. At first hearing it may appear absurd, but the point is a sound one. When catastrophe strikes, sometimes the best long term solution is to accept severe initial damage and survive rather than to attempt to avoid it with a more serious result. This is directly applicable to many software development projects, which always seem to be rudderless in a sea of potential disasters.