Resumen
This article discusses how to judge the usefulness of aspect-oriented programming (AOP). There are two basic techniques for assessing a programming technology namely experiments and case studies. Experiments provide an opportunity for direct comparison, allowing researchers to investigate how the technology performs in detailed ways. Case studies take many forms, from small-scale comparisons to longitudinal studies of the technology in action. The authors conducted two semicontrolled experiments to investigate whether AOP, as embodied in AspectJ, made it easier to develop and maintain certain kinds of application code. One of the experiments showed that when locating a fault involved reasoning within a single class or an aspect, programming teams using AspectJ were able to correct a seeded program fault faster than the Java programming language teams. Based on the results of the experiments and case studies conducted by the authors, AOP shows promise. The authors have determined particular situations where AOP benefits developers, and have characterized aspects of how AOP has worked, presenting design and process guidelines they have found helpful in developing AOP systems.