The vision of ubiquitous computing is as exiting as it is challenging. Computing devices came an amazingly long way in the past few years, and plentiful connectivity is becoming commonplace. However, the increasing numbers and richness of devices is a mixed blessing for users. Users today carry a heavy burden in finding and configuring applications for their computer-supported activities, making information available and consistent across different devices, and fixing things when they break.
Ideally, a ubiquitous computing infrastructure would offer personalized support to a user's activities at any location, taking into account preferences with respect to quality of service and concerns with respect to privacy. Such infrastructure would enable users to take full advantage of all the computing capabilities at each location, much like people today can take advantage of the furniture in different spaces. Furthermore, it would automatically adapt to dynamic changes both in computing resources and in the requirements associated with the user's activities.
Still unclear is how software should be organized to deliver this vision, and which conceptual metaphors should be offered to the users of such software.
This work aims towards this vision, focusing on office-like environments. The key idea is the introduction of a software layer above applications. This new layer exploits models of user needs and preferences at runtime to find the best match between those needs and the computing capabilities at each location. Such optimality is maintained in the face of dynamic changes.
I developed Prism, a task manager that supports user mobility (demo video).
Prism’s visible components are:
· Dashboard, lists the currently pending tasks, video tutorial*
· Lamp, a task browser, video tutorial*
· Focus, supports the incremental definition of tasks, video tutorial*
If you have an installation of Prism, you may want to refer to the logistics tutorial (you may want to watch this video in full screen mode).
Task-based Adaptation for Ubiquitous Computing, IEEE SMC 2006
Conference & Workshop Papers
Task-based Self-adaptation, WOSS 2004