Quality-Aware Mashup Composition: Issues, Techniques and Tools

Web mashups are a new generation of applications based on the composition of ready-to-use, heterogeneous components. In different contexts, ranging from the consumer Web to Enterprise systems, the potential of this new technology is to make users evolve from passive receivers of applications to actors actively involved in the creation of their artifacts, thus accommodating the inherent variability of the users' needs. Current advances in mashup technologies are good candidates to satisfy this requirement. However, some issues are still largely unexplored. In particular, quality issues specific for this class of applications, and the way they can guide the users in the identification of adequate components and composition patterns, are neglected. This paper discusses quality dimensions that can capture the intrinsic quality of mashup components, as well as the components' capacity to maximize the quality and the user-perceived value of the overall composition. It also proposes an assisted composition process in which quality becomes the driver for recommending to the users how to complete mashups, based on the integration of quality assessment and recommendation techniques within a tool for mashup development.