ReWiRe: creating interactive pervasive systems that cope with changing environments by rewiring

Abstract

The increasing complexity of pervasive computing environments puts the current software development methods to the test. There is a large variation in different types of hardware that need to be addressed. Besides, there is no guarantee the environment does not evolve, making the software developed for the initial environment deprecated and in need for updates or reconfiguration. Software deployed in such an environment should be sufficiently dynamic to cope with new environment configurations, even while the system is in use. This goes beyond coping with new contexts of use and building context-aware systems: while most approaches are mainly focused on how the software behavior adapts according to the changing context in a fixed environment, our approach, ReWiRe, allows the environment configuration to change over time.

Combining aspect-oriented modeling with property-based reasoning to improve user interface adaptation

“…ReWiRe is a framework dedicated to the dynamic adaptation of interactive systems [26]. As in our approach, ReWiRe's architecture uses a component-based system that facilitates the (de-)activation of the system's components.…”

Put the User in Control: Ontology-Driven Meta-level Interaction for Pervasive Environments

“…The proposed approach was successfully applied in several practical cases, e.g. a pervasive paint application and an arcade game [12].…”

“…For example, when a UI is rendered to enable a user task, proxies to interact with the task's related services are automatically constructed and passed to the UI component. For more information on the technical details of the grounding process, we refer to [12].…”

Improving intelligibility and control in Ubicomp

“…As a first step, I extended the existing ubicomp framework ReWiRe [19] to allow systems built with ReWiRe to reason about the causes and consequences of system and user actions. As ReWiRe employs a simple rule-based behavior model it is a good candidate for exploring different interaction techniques for intelligibility and control.…”