Summary

Software testing today seems to have four main problems.

First, changes in the way that software is developed have eroded the value of software testing. New development methodologies are developed iteratively through trial and error. These methods typically don't produce specifications for the tester to test against, so testers are left to hunt for bugs.

Second, in recent years, software development has been driven by entrepreneurial pressures, schedule, and constantly evolving product definition. For these and other reasons, management is not always convinced that testing is necessary or worthwhile. In many types of development effort, testing by an independent test group is not believed to be worthwhile or to add value.

Third, there are few trained testers using formal methods and metrics, and most of the software testers out there are just passing through testing on their way to some other (more promising and rewarding) career. As a result, the test effort is not producing the kind of high-quality results that help demonstrate how the test effort improves the quality of the product and the bottom line.

Finally, the most important, and visible, software quality improvements in the past several years have not come as a result of testing, but of standardization driven by the Internet. This increased use of standards has had a profound effect on the quality of the final product and on the ability of multiple software systems to interact and interoperate. It has also removed the need for extensive low-level testing in these areas, so the demand for senior technical testers is decreasing.

In the end it comes down to the question "Does your test effort add enough value to make it worthwhile?" It used to be assumed that the answer to this question was yes. Today, however, this is not true. A test effort that just finds bugs is not enough. Testers must be able to demonstrate that their effort is adding value to the bottom line. There are many ways to measure that value, but you have to be measuring all along in order to demonstrate the value added. Managers and auditors, and just about anyone looking at the bottom line, expect to see a real, measurable demonstration that every part of the development of a product is contributing. This book shows you many ways to show this value. In the next chapter, we discuss some of the problems with traditional quality assurance in today's software development environment.