Description:
If patterns are good ideas that can be re-applied to new situations, AntiPatterns: Refactoring Software, Architectures, and Projects in Crisis looks at what goes wrong in software development, time and time again. This entertaining and often enlightening text defines what seasoned developers have long suspected: despite advances in software engineering, most software projects still fail to meet expectations--and about a third are cancelled altogether. The authors of AntiPatterns draw on extensive industry experience, their own and others, to help define what's wrong with software development today. They outline reasons why problem patterns develop (such as sloth, avarice, and greed) and proceed to outline several dozen patterns that can give you headaches or worse. Their deadliest hit list begins with the Blob, where one object does most of the work in a project, and Continuous Obsolescence, where technology changes so quickly that developers can't keep up. Some of the more entertaining antipatterns include the Poltergeist (where do-nothing classes add unnecessary overhead), the Boat Anchor (a white elephant piece of hardware or software bought at great cost) and the Golden Hammer (a single technology that is used for every conceivable programming problem). The authors then proceed to define antipatterns oriented toward management problems with software (including Death by Planning and Project Mismanagement, along with several miniature antipatterns, that help define why so many software projects are late and overbudget). The authors use several big vendors' technologies as examples of today's antipatterns. Luckily, they suggest ways to overcome antipatterns and improve software productivity in "refactored solutions" that can overcome some of these obstacles. However, this is a realistic book, a mix of "Dilbert" and software engineering. A clever antidote to getting too optimistic about software development, AntiPatterns should be required reading for any manager facing a large-scale development project. --Richard Dragan
| 
