Pablo Trinidad Martín Arroyo

Mass Customisation Along Lifecycle of Autonomic Homes

Autonomic homes adapt themselves to give the user the best possible experience of the services they provide. They dynamically adapt its behavior at run-time in response to changing conditions in end-user needs and the surrounding environment devices. From the development point of view, producing and maintaining a large amount of autonomic homes need an affordable solution such as dynamic software product lines (DSPL). DSPL produce a set of products that share features and have the ability of reconfiguring at runtime. Since users maintain and modify their preferences in opportunistic and improvisational ways, an autonomic home must evolve in time according to user expectations. Current DSPL architectures implement the ability of reconfiguring a product but ignore user preferences. We present an extension to our DSPL architecture to incorporate user preferences so user customisation of autonomic homes is not limited to installation time but all along the lifetime.

Configurable Feature Models

Feature models represent all the products that can be built under a variability-intensive system such as a software product line, but they are not fully configurable. There exist no explicit effort in defining configuration models that enable making decisions on attributes and cardinalities in feature models that use these artefacts. In this paper we present configurable feature models as an evolution from feature models that integrate configuration models within, improving the configurability of variability-intensive systems.

FaMa Abductive: una herramienta para explicaciones de errores en modelos de características

Resumen La diagnosis de errores es una asignatura pendiente de gran parte de las herramientas de análisis de modelos de caracteŕısticas. Realizar este proceso manualmente no es viable con modelos de tamaño medio/grande, e incluso tampoco para modelos pequeños, pues la mayoŕıa de errores son extremadamente dif́ıciles de detectar y más aún de explicar. En esta demo presentamos FaMa Abductive, una extensión de la herramienta FaMa, encargada de la diagnosis automática en modelos de caracteŕısticas. Esta herramienta provee detección y explicaciones para errores en modelos de caracteŕısticas, tanto básicos como extendidos, y también para productos erróneos en modelos de caracteŕısticas básicos.

Tool Supported Error Detection and Explanations on Feature Models

Jesus García-Galán Dept. Computer Languages and Systems University of Seville Avda. Reina Mercedes s/n, 41012, Seville, Spain jegalan@us.es Pablo Trinidad Dept. Computer Languages and Systems University of Seville Avda. Reina Mercedes s/n, 41012, Seville, Spain ptrinidad@us.es José A. Galindo Dept. Computer Languages and Systems University of Seville Avda. Reina Mercedes s/n, 41012, Seville, Spain jagalindo@us.es