Avaliação do produto como ferramenta para melhoria de software

Abstract

Since the 1950s we have seen a considerable increase in demand for computer-based systems. But the increased demand not only meant a larger amount of systems to develop, but also an increase in the complexity of these systems. Whether people previously worked with reduced-scope, single-user, desktop systems, we now have systems that have to: deal multiple concurrent users, deliver web-enabled access, fit and adapt to different types of access devices and other new technologies, integrate with databases, among others. However, regardless of the size or complexity of these systems, the demand for quality is always a constant. Among many possible definitions of quality, we can say that it is, basically, to be adherent to requirements. One of the problems that software engineering need to deal with today is, precisely, how to measure the quality of a software product satisfactorily. Nowadays, there are several proposals in the area of standards and of software quality assessment, but mostly appear as generic methods and, hence, do not properly evaluate the quality from the viewpoint of the application domain. In order to establish a specialized method, one should take into account the whole software specification. Aware of the importance of software product evaluation as a supporting agent of system's evolution, we propose in this work a process for continuous evaluation of a software product, using concepts already established and standardized, within a scenario previously selected and, thus, taking into account the particularities of that system. Moreover, we performed an execution of the proposed process, using a quality model also built for the needs identified in that scenario. With this, we hope to contribute to the evolution and improvement of that deployed software system.