Toolkits for Ubiquitous Computing

Weiser’s vision of ubiquitous computing assumed that devices would be able to talk to each other and work together seamlessly [1]. Want [2] surveys exploratory research in achieving device-to-device communication in various systems to achieve seamless computing and identifies that early systems required a central sensing system (e.g., GaTech’s Aware Home and Olivetti’s Active Badge). Korteum et al. [3] consider the problem at a finer level of granularity and do not assume that a central sensing system exists. Instead, they introduce a type of device that can be practically affixed to existing computational devices via USB dongle that gives relative 2D positional information. They are not as concerned with the interactions that are possible, but with the hardware and toolkit support that is necessary in order to investigate practical ubiquitous computing system.

I am pleased that concepts such as subscription to relationship events (e.g., close together) in Korteum el al. [3] Relate toolkit design has been incorporated into the 3D version, Proximity Toolkit [4]. Both toolkits capture low-level data and formulate high-level constructs that are useful for rapid prototyping and development. The types of high-level information they return are not necessarily useful, e.g., the absolute distances between devices in Korteum et al. [3] were found to be less relevant in the visualization than the relative distance between devices.

The toolkit introduced by Korteum et al. [3] uses a bottleneck, the USB dongle. Without the USB dongles, there is no ability to prototype proximity-based interactions. The design of Proximity Toolkit envisions that there could be multiple proximity-based inputs, some more accurate than others, by not assuming a particular technology, but so far it supports Vicon system primarily.

What do we have to do to make such proximity systems available to more researchers? I think there has to be a change in policies for conferences publishing toolkit papers. Toolkit papers are only useful it other researchers can access the toolkit. The conferences might require a version of the toolkit, open source if possible, to be submitted and distributed for others to access and expand upon. I think there are other possibilities and leave this for open discussion.

1. Weiser, M. The computer for the 21st Century. Scientific American. 94-110, September. (1991)

2. Want, R. An Introduction to Ubiquitous Computing. In Ubiquitous Computing Fundamentals, Krum, J. (Editor), CRC Press. 2010.

3. Kortuem, G., Kray, C., and Gellersen, H. Sensing and visualizing spatial relations of mobile devices. Proceedings of the 18th annual ACM symposium on User interface software and technology, ACM New York, NY, USA (2005), 93-102.

4. Diaz-Marino, R. and Greenberg, S. (2010) The Proximity Toolkit and ViconFace: The Video. In Video Showcase, DVD Proceedings of the ACM Conference on Human Factors in Computing Systems - ACM CHI'10. (Atlanta, Georgia), ACM Press, 5 pages, April 10-15. Video and paper, demonstrated live at CHI.