Researchers seeking alternatives to traditional desktop computers have begun exploring the potential collaborative benefits of digital tabletop displays. However, there are still many open issues related to the design of collaborative tabletop interfaces, such as whether these systems should automatically orient workspace items or enforce ownership of workspace content. Understanding the natural interaction practices that people use during tabletop collaboration with traditional media (e.g., pen and paper) can help to address these issues. Interfaces that are modeled on these practices will have the additional advantage of supporting the interaction skills people have developed over years of collaborating at traditional tables. To gain a deeper understanding of these interaction practices we conducted two observational studies of traditional tabletop collaboration in both casual and formal settings. Our results reveal that collaborators use three types of tabletop territories to help coordinate their interactions within the shared tabletop workspace: personal, group, and storage territories. Findings from a spatial analysis of collaborators' tabletop interactions reveal important properties of these tabletop territories. In order to provide a comprehensive picture of the role of tabletop territoriality in collaboration, we conclude with a synthesis of our findings and previous research findings and with several relevant design implications.

No abstract

Previous research has shown that rotation and orientation of items plays three major roles during collaboration: comprehension, coordination and communication. Based on these roles of orientation and advice from kinesiology research, we have designed the Rotate'N Translate (RNT) interaction mechanism, which provides integrated control of rotation and translation using only a single touch-point for input. We present an empirical evaluation comparing RNT to a common rotation mechanism that separates control of rotation and translation. Results of this study indicate RNT is more efficient than the separate mechanism and better supports the comprehension, coordination and communication roles of orientation.

Abstract. In order to support co-located collaboration, many researchers are now investigating how to effectively augment tabletops with electronic displays. As far back as 1988, orientation was recognized as a significant human factors issue that must be addressed by electronic tabletop designers. As with traditional tables, when people stand or sit at different positions around a horizontal display they will be viewing the contents from different angles. One common solution to this problem is to have the software reorient objects so that a given individual can view them 'right way up.' Yet is this the best approach? If not, how do people actually use orientation on tables? To answer these questions, we conducted an observational study of collaborative activity on a traditional table. Our results show that the strategy of reorienting objects to a person's view is overly simplistic: while important, it is an incomplete view of how people exploit their ability to reorient objects. Orientation proves critical in how individuals comprehend information, how collaborators coordinate their actions, and how they mediate communication. The coordinating role of orientation is evident in how people establish personal and group spaces and how they signal ownership of objects. In terms of communication, orientation is useful in initiating communicative exchanges and in continuing to speak to individuals about particular objects and work patterns as collaboration progresses. The three roles of orientation have significant implications for the design of tabletop software and the assessment of existing tabletop systems.

No abstract