We describe a prototype virtual reality teleconferencing system called MASSIVE which has been developed as part of our on-going research into collaborative virtual environments. This system allows multiple users to communicate using arbitrary combinations of audio, graphics, and text media over local and wide area networks. Communication is controlled by a so-called spatial model of interaction so that one user's perception of another user is sensitive to their relative positions and orientations. The key concept in this spatial model is the (quantitative) awareness which one object has of another. This is controlled by the observing object's focus and the observed object's nimbus , which describe regions of interest and projection, respectively. Each object's aura defines the total region within which it interacts. This is applied independently in each medium. The system (and the spatial model which it implements) is intended to provide a flexible and natural environment for the spatial mediation of conversation. The model also provides a basis for scaling to relatively large numbers of users. Our design goals include supporting heterogeneity, scalability, spatial mediation, balance of power, and multiple concurrent meetings; MASSIVE meets all of these goals. Our initial experiences show the importance of audio in collaborative VR, and they raise issues about field of view for graphical users, speed of navigation, quality of embodiment, varying perceptions of space, and scalability.

This paper introduces a spatial model,of interaction which aims to support groups of people ,in using ,their natural communication ,skills in distributed virtual environments. First, we outline our motivation for undertaking this work in terms,of the social significance of space,in supporting,co-operative work. Next, we summarise the model's main concepts, namely aura, awareness, focus, nimbus, adapters and boundaries. Following this, we describe routes to implementing the model. The first is an approach suited to realising the model as an application of existing VR platforms. The second is ageneral implementation of the model, based on its own specially tailored distributed architecture. Finally, we present our initial observations from use in the laboratory setting and outline issues for future work.To appear in Proceedings VRST'94, Singapore, August 1994, ACM Press.

We propose an approach to creating shared mixed realities based on the construction of transparent boundaries between real and virtual spaces. First, we introduce a taxonomy that classifies current approaches to shared spaces according to the three dimensions of transportation, artificiality, and spatiality. Second, we discuss our experience of staging a poetry performance simultaneously within real and virtual theaters. This demonstrates the complexities involved in establishing social interaction between real and virtual spaces and motivates the development of a systematic approach to mixing realities. Third, we introduce and demonstrate the technique of mixed-reality boundaries as a way of joining real and virtual spaces together in order to address some of these problems.

We explore the approach of performance-led research in the wild in which artists drive the creation of novel performances with the support of HCI researchers that are then deployed and studied at public performance in cultural settings such as galleries, festivals and on the city streets. We motivate the approach and then describe how it consists of three distinct activities -- practice, studies and theory -- that are interleaved in complex ways through nine different relationships. We present a historical account of how the approach has evolved over a fifteen-year period, charting the evolution of a complex web of projects, papers, and relationships between them. We articulate the challenges of pursuing each activity as well as overarching challenges of balancing artistic and research interests, flexible management of relationships, and finally ethics.