A new book on software testing isn't quite perfect, and it isn't just about software, but it is a worthy read for anyone managing projects that involve testing. The heart of the book describes a testing process in terms of Satir’s four-stage interaction model: gathering information, assigning meaning, determining significance and deciding what (if anything) to do.
Like most of Jerry Weinberg's books, his new one, Perfect Software and Other Illusions About Testing, isn't really about software at all. It's about people: people who develop things that need to be tested, people who perform tests, people fix the problems found by testing, and people who manage projects that involve testing. And while the world of software development provides Weinberg with examples and stories to make his points, the lessons of the book apply to just about anything that must be tested.
Weinberg begins with the most fundamental questions: why do we test? What do we hope to find by testing? What will we do with the information we get from testing? What should we test? What should we perhaps not test? How will we know when it's time to stop testing? Where do you draw the line between testing and fixing? How, and how much, should you test the testers and the whole process of testing?
Weinberg takes time to consider how people managing testing can answer these questions in ways that