Browse Books

Reviewer: James M. Perry

This is a collection of papers from the first annual Conference on Extreme Programming and Flexible Processes in Software Engineering, held in Italy in June 2000. Through careful selection and organization, this collection of 33 papers provides an excellent overview of the issues and discussions on the new development methodology, Extreme Programming. The collection of papers cover a broad and diverse set of topics, grouped into parts: Part 1: principles of XP Part 2: some methodology and process issues for XP, including topics on reference architecture, frameworks, project organization and large scale software development Part 3: XP in relationship to the Unified Modeling Language (UML) and the Unified Software Development Process (USDP), maintenance and patterns Part 4: XP key practices, including pair programming, unit testing and refactoring Part 5: tools to support XP, such as refactoring browser, version control, collaboration and communication and automated testing Part 6: experiences in transitioning an existing project to XP, adopting XP, XP in a research and development laboratory Part 7: other topics, such as on learning XP, XP in maintenance, C++ design support for XP, tracking via variability, software agents and product lines. Given that the papers are written by different authors, the collection is surprisingly unified. Each chapter is a separate paper, beginning with a concise abstract and ending with conclusion, which nicely summarizes the paper, and a reference section. Each paper deals with a specific concept which is intrinsic to XP, or which can be related to XP. The concepts are not discussed in depth, but enough information is given to identify the concept and introduce it to the reader. The reference section, as a result, is especially important for readers who wish to follow-up these introductions in greater detail. In addition, the book has a useful index that spans all the papers. While the range of topics is broad, some topics are not covered. For example, version control is covered, but formal or baseline control is not; unit testing is covered, but integration testing and system testing are not; except for a brief discussion of tracking variation, measures or metrics are not covered; software quality is not covered; pair programmer continuous reviews are covered, but more formal peer reviews are not covered. While these are not mainstream topics in XP and other flexible methods, they are important in traditional development methodologies for control, verification, and visibility, and they motivate questions of how they compare to XP practices and strategies. Many of the articles are of interest outside of the XP community. There are ideas on how to lighten the overhead of heavyweight processes, on technology transition, on software reuse, patterns, programming teams, and unit testing. As an example, the paper “Unit Testing in a Java Project” by Peter Gassmann is a unit test treatise covering test organization and tool support, and introducing unit tests in a project, what to test, how to test, when to write tests, when to run the tests and the benefits of unit testing. Thus, the book is of interest to those who do not subscribe to XP, but who are open to new ideas that can improve traditional development methods. There is a paper for almost everyone. By reading the papers’ abstracts, readers can select papers pertaining to their interests, follow a thread by selecting one paper from each Part, read a Part in its entirety, or read all the papers sequentially. For example, if one were interested in the application of XP in industry, an appropriate subset and thread of papers might be: “Is Design Dead__ __” by Martin Fowler, “XP and Large Distributed Software Projects” by Even-Andre Karlsson and Lars-Goran Andersson, “XP Inside the Trojan Horse: Refactoring the Unified Software Development Process” by Jutta Eckstein and Rolf F. Katzenberger, the paper on unit testing mentioned in the preceding paragraph, “Team Streams: Extreme Team Support” by Jim des Rivieres, Erich Gamma, Kai-Uwe Matzel, Ivan Moore, Angre Weinand, and John Wiegand, “The VCAPS Project: An Example of Transitioning to XP” by Don Wells and Trish Buckley, and “Legacy to the Extreme” by Arie van Deursen, Tobias Juipers, and Leon Moonen. New software development methodologies have emerged in response to problems encountered using existing methodologies on software projects. New approaches usually focus on a particular technology, for example programming and design, or management processes. XP is itself a response to the problems of rapid change of requirements, complexity, and late feedback of progress. XP addresses these problems with a focus that spans technologies for user involvement, incremental change and frequent testing, simplicity, and pair-wise programming, all in the eXtreme. XP has adopted existing solutions by combining them in new ways, with extreme emphasis. Extreme Programming Examined is a nice diverse blend of background, techniques, tools, practical experiences, and speculation. Online Computing Reviews Service