Testing Extreme Programming

If you write software, you must test what you have done. Extreme Programming is unique in the focus it places on the testing aspects of development. Great book for QA Agilists.

This book explains, in detail, the value of having a tester role on an XP project. Much of what is written on XP focuses on the role of developers in writing tests, and this book will help you understand the value that testing professionals can add to an XP, or other agile, project. The function of a tester changes when a team applies XP; In the spirt of lean software development, this book shows you how to apply the skills of testers in a new environment. One minor flaw in the book is that it is targeted at testers. While testers need to understand how to apply their skills to an XP environment, others such as Coaches, programmers and project managers also need to understand the value of testers. Get this book if you are a tester, developer, coach, or project manager to understand how testers can improve your XP project.

While I yield to no one in recognizing the value of testing software, my first reaction to reading the title of this book was one of skepticism. One of the fundamental principles of extreme programming (XP), is that the software is developed in small increments, each of which must pass a unit test before the next change is made. In fact, in many cases the unit test is written by the developers before they write the code. These incremental tests are also carried out by the two-person coding team, so it seemed strange to be reading about testing XP. From the title and blurbs on the covers, it was a natural assumption that the focus would be on testing other than that done by the development teams. After reading the book, that skepticism has largely gone, although I do possess some residual doubts about XP and how it scales. The basic point is that programmers are very good at testing their code at the unit level, but weak when asked to verify it at the system level. I agree with the authors that there should be a dedicated tester who examines the code at a level higher than the unit. However, I am also of the opinion that this is a confirmation of the doubt about XP expressed by so many observers, namely that it does not scale up to large projects well. The testers that they are proposing are more in the realm of a manager responsible for testing rather than a tester. This is of course very sensible. Once the programmers start producing code tested at the unit level and the integration process begins, someone must be responsible for the smooth flow and testing of the integration. This is also the level where the ever-present customer, another fundamental principle of XP, really sees the functionality of the code for the first time. While XP proponents speak a great deal about having the customer at the side of the coding team, realism dictates that they will generally be restricted from that level. Only the most technically sophisticated customer will be able to glean any useful information from most of the unit tests that will be performed. This is where the additional layer of the test manager is of use. By creating and demonstrating the higher level tests, the test manager can give the customer information that they will understand and can respond to. The authors also put forward a very controversial statement, "No manual tests. All acceptance tests on an Extreme Programming project must be automated." While I am in general agreement with the principle that tests should be automated for easy repetition at each level, the reality is that nearly every use of words such as { "no", "never", "all"} is too extreme. Especially when you are describing something as subjective as the behavior of computer programs and the human response to them. How one can automate the response of a customer to the appearance of a GUI interface is something I do not yet understand, and this is mentioned, but not examined in the book. One very positive aspect of the b

What a great book on Testing in the Xtreme Programming Environment! I expect this book to become the Bible for test engineers and customers who are engaged in formal testing on an Xtreme Programming project. Lisa and Tip walk us through some introductory theory, and then provide a detailed case-study to illustrate their methods. Though appearing somewhat complex to a non-programmer, their chapters 21 through 25 show the important principles of how to automate your acceptance test.Though I haven't yet had the experience of working on an Xtreme Programming project, I do have 30 years of hardware, systems, and software testing experience. Prior to reading this book, I read Kent Beck's "eXtreme Programming eXplained", so I had an introduction to Xtreme Programming.Once again, this is a great book. My welcome to 2 new authors; I hope we see more books from them in the future.

This book is fantastic. I'm only a tester to the extent that everyone doing XP is a tester but I found many valuable things in this book. There are excellent chapters about how to estimate and plan releases and iterations, including excellent tips on finding hidden assumptions in your stories. I like the book so much that I just bought a copy for a tester who works for me and told him to go home, read it, and not come back until he has so that we can discuss how to put some of the book's advice into practice. Everyone working on an agile project (not just XP and not just testers) will benefit from this book. It is one of the best-written and most useful in the entire series.