Exploring Casual Tabletop Interactions

Abstract

Although there are many different types of tabletop displays in the literature, there are very few fundamentals known about what tabletop interfaces should look like or should behave. In order to obtain a better understanding of these questions, it seemed appropriate to back up a little from the interface design, and look at tasks and environments that we would like to support. Thus, several observational studies of traditional tabletop collaboration were undertaken to inform the design of collaborative tabletop interfaces. To get a broad understanding of tabletop collaboration in a in a variety of settings and tasks, these studies involved both informal and formal collaboration. This chapter describes the informal collaboration study, which involved participants playing tabletop games in a drop-in setting, inspired by kiosks setup in many science museums.

The purpose of these studies was to gain a better understanding of how people use the space provided by a tabletop surface and how they use artifacts on a table during collaborative tabletop activities. This information can help to develop effective tabletop display technology to support face-to-face collaboration. There has been a variety of tabletop display systems developed in regent years (Durbin et al., 1998; Rekimoto & Saitoh, 1999; Streitz et al., 1999; Vernier et al., 2002; Wellner, 1993), but there has been little investigation of the usability of these systems for collaborative activities. More importantly, most of the interaction styles used in these systems have evolved from interaction styles used in existing vertical displays systems, such as the typical desktop system running Windows, Mac, or Unix.

Two disadvantages of using these interaction styles to support collaboration on a tabletop display include: 1) the interaction styles have been optimized for use on a vertical display surface, and 2) they have been optimized for single-user interaction with the computer 9most computer-supported collaboration occurs over a network, with each collaborator interacting with their own computer). Vertical-display interfaces have a fixed orientation of the display artifacts (e.g. windows, dialog boxes) that has an obvious �top� and �bottom�. This does not allow easy viewing for multiple people who are seated at various sides of a tabletop display. Single-user interfaces usually have only one of each type of input device (e.g., one mouse, one keyboard). This does not allow concurrent interaction for multiple users, which collaboration researchers have shown occurs when people interact in non-technology environments (using pen and paper) (Tang, 1991) and in technology environments that provide support for multi-user, concurrent interaction (Scott et al., 2000).

Understanding how people interact with artifacts and with the space on the table in a non-technology environment can help us develop tabletop display systems that support natural collaborative behavior around a horizontal surface, instead of trying to evolve technology that has been optimized for other uses. The first study employed an ethnographic-style methodology, where hand-written observations were taken while people performed a variety of simple, non-technological, collaborative tabletop activities in a casual, drop-in activity area.