Aspect-Oriented Programming evaluated: A Study on the Impact that Aspect-Oriented Programming can have on Software Development Productivity

Aspect-oriented-programming is a relatively new technique that has evolved on top of the already well-established approach of object-oriented programming. When it is used correctly, it promises to remove many redundant parts of a code that appear repeatedly in an application, essentially untangling the original code. Thus, it can lead to a cleaner, more separated software design, to greater modularity and maintainability. Time-savings in software engineering can also be huge cost-savings, and anything that increases software quality is a welcome sight in an industry that in parts, still suffers from a bad reputation concerning the quality of its products. But, used in the wrong way or with insufficient knowledge, aspect-oriented programming can result in the opposite. Unintended side effects may introduce new bugs into an application, and therefore, it may be just one of the many negative consequences. In any case, there is a huge lack of empirical research on the subject, and in software engineering in general. Due to that fact, the question arises in how far hard facts can be drawn. This book deals with part of that problem by evaluating the aspect-oriented approach in a carefully designed and controlled experiment. On the basis of different tasks that were solved by participants of this study, the aspect-oriented approach (represented by AspectJ) is compared with a plain object-oriented approach (represented by Java). The book starts with an introduction to the topic, and further, it provides on the one hand, the survey's motivation and on the other hand, some background information. A short chapter on aspect-oriented programming and empirical research may help readers who are unfamiliar with any of the subjects. Then, the survey itself is described in detail, i.e. its design, its implementation, and its evaluation, followed by a thorough discussion of the results. And the answer to the question 'Can aspect-orientated programming keep its promise in productivity?' is given.