Collaborative Virtual Environments

In the late 1980s Virtual Reality (VR) burst onto the public stage propelled by a wave of media interest and related science fiction novels such as Neuromancer by William Gibson (Gibson, 1989). VR promised to revolutionize the way in which we experience and interact with computers, and research into the field mushroomed. More recently, the hype surrounding VR has died down and, although it is receiving less public attention, serious work is continuing with the aim of producing useful and usable technology. At the centre of current work related to VR is the field of Collaborative Virtual Environments (CVEs). This field has as its goal the provision of new, more effective means of using computers as tools for communication and information sharing with others. Many CVE systems have been constructed. Some of these are desktop systems and applications; but large public virtual spaces have also been constructed (such as Alpha World at http://​www.​activeworlds.​com/​; see Chapter 15). CVEs are also being used to experiment with new forms of art and interactive television (Benford et al., 1997a,b; Benford et al, 2000a).

This chapter is about extending CVEs to include more people and more objects in more media. Fundamentally the story goes like this.

According to Professor Fred Brooks, virtual environment (VE) research has reached a point where it “barely works” (Brooks, 1999). We interpret this to mean that, although the technology has advanced to a point where many things are possible, there are still inadequacies which prevent its application to real-world problems. This difficulty of getting virtual reality (VR) to work adequately is due to a wide range of things: some are hardware-related, some depend on better algorithms and techniques, and yet others are related to human factors issues. This very diversity has led to some fragmentation in research: different groups study focused, tractable problems; solutions for one problem do not necessarily integrate easily with others, and technological limitations often frustrate attempts to scale up the results for larger real-world tasks. Gluing together the solutions for individual components of the problem is not guaranteed to work, and we can see evidence that the overall task is hard from the lack of convincing industrial-strength examples. Finding a solution for a real-world problem requires that diverse ideas and solutions be combined in some way. Before VR can move from a research-based subject into mainstream applications, ways must be found to integrate ideas from the contributing disciplines into a coherent architectural framework, designed to support systematic application development. Our own work, which is briefly touched upon in this chapter, is concerned with deriving such a system architecture.

Collaborative Virtual Environments, like almost all distributed systems, are complex and encompass a broad range of interrelated concerns. In this chapter we focus on the network and communication requirements of CVEs in relation to several aspects of their design, implementation and use. We cannot cover the relationships between these various aspects comprehensively in this chapter; rather we will give a framework and a number of key illustrations which will equip the reader with a general understanding of the issues and the typical patterns of influence which are embodied in CVEs.

In this chapter we seek to develop an understanding of the nature of presence in virtual environments from a real-world exploration of first, how presence is manifest in practice within virtual environments and second, the means by which everyday users experience virtual environments. The ethnographic study presented in this chapter is part of a multidisciplinary approach to the design of online large-scale collaborative virtual environments (CVEs) undertaken by the authors in a range of projects. Over the last few years a large number of systems to support online virtual worlds and CVEs have emerged. These include 3D environments: MASSIVE (Greenhalgh and Benford, 1995), DIVE (Dix, 1997), AlphaWorlds, Blaxxun, VWWW; 2D graphical systems: ThePalace; and textual environments: MUDs (Edwards, 1994).

In this chapter, we focus on collaborative virtual reality systems that use a combination of 3D graphics and audio to enable people to interact within a virtual setting and to discuss, fashion and manipulate their common environment. The emergence of these systems introduces unique opportunities to develop new sites of sociality and to support distributed collaborative work and interaction. Indeed, as 3D visualizations and single-user VR technologies become increasingly adopted within numerous industrial and entertainment domains (Schroeder, 1996; Stanney et al., 1998), there are heightened opportunities for associated collaborative applications to emerge (Biocca and Levy, 1995). Most successful thus far have been the entertainment applications, such as online gaming, inhabited TV, artistic installations and museum exhibits (see Benford et al., 1997a,b, 1999a; Greenhalgh et al., 1999b,c; and Chapter 5 of this volume). However, here we are particularly concerned with the potential for collaborative VR to provide support for remote working - a relatively under-explored area.

There is today a spate of Collaborative Virtual Environments (CVEs) on the net (http://​www.​digitalspace.​com/​avatars/​ indexes them), similar to the Metaverse in Neal Stephenson’s Snow Crash (Stephenson, 1992). Some of these allow the users to construct buildings in the world, though in the majority the environment is fixed. These buildings serve as meeting points and atmosphere creators, but do not have any other meaning. Information can be presented in these spaces as imported objects or as placards with text and images placed in the environment.

People in a group play various roles. Manager, intern, department chair, guest speaker, the woman who knows how the fax machine works, the guy who got the pizza, the person speaking now, the person capturing this on the whiteboard; roles can be formal or informal, long-lived or ephemeral. They can be bestowed by ceremony, or assumed with the invisibility of the taken-for-granted. But it is the interaction of minds that creates and sustains a role: a role is a human construct. Even roles like “mother” or “father”, which would seem to demand certain biological characteristics, cannot be defined at any depth without appeal to social convention. The largely hands-off approach taken by nature seems to work fairly well for groups of humans: when there is confusion or concern about who is playing what role, it is generally good for a group to work it out as a social issue. If roles were encoded solely in human biology or in physical law, life might be simpler but certainly more stulfiying.

We describe our experiences of creating and using a wide variety of techniques and applications to support collaboration in virtual environments for different activities and user groups. Our aim is to offer plentiful and rich possibilities for interaction across and between both real and virtual environments. This is informed through ongoing activities and projects, starting in the early 1990s, that have provided initial discoveries that point toward what is necessary to support interaction in a CVE. In the early stages, the focus of the work was on creating a programmable CVE, and the DIVE system (Frécon and Stenius, 1998) was born. Starting off as a simple, collaborative 3D graphical environment, the sophistication and functionality of DIVE have progressed, and so have the scenarios in which it is used.

Collaboration, and increasingly multidisciplinary collaboration across distances, is a fundamental and strategic component of the scientific research process. The importance of collaboration in the scientific research process has long been recognized by the National Institutes of Health (NIH) in the USA. Its current policy is to fund the development or purchase of specialized scientific instruments in (not-for- profit) research labs across the USA and to help fund scientists to travel to those labs to collaborate and conduct scientific experiments using these specialized scientific instruments.

In this chapter we will discuss our work to enable effective collaboration between remote participants within immersive virtual reality environments. We will discuss several of the environments we have created, the lessons we have learned and the issues that are driving our current research. Our focus is on high-quality collaboration among a handful of participants connected by high-speed, high-bandwidth national and international networks in areas such as scientific visualization, education, and design. Sharing these virtual environments with your remote collaborators as well as the topic of your collaboration may be better than collaborating with them in person, as you can now collaborate while standing inside your data, sharing the context of your discussions.

Virtuality need not be a prison. It can be the raft, the ladder, the transitional space, the moratorium that eventually is discarded in order to reach greater freedom. We don’t have to reject life on the screen, but we don’t have to treat it as an alternative life either. Like the anthropologist returning home from a foreign culture, the voyager in virtuality can return to a real world better equipped to understand its artifices. (Turkle, 1998, p. xvii)

Waterfall Glen is a MOO, a form of MUD or “multi-user domain”. MUDs are immersive virtual environments - one enters the MUD environment and conversations take place within it. Traditionally MUDs have been used for multi-player Internet gaming and for social chat, and have not been used extensively to support work-related activities. However, MUDs have been gaining attention as potential environments for supporting collaborative learning and cooperative work. Recent publications describe the use of MUDs for supporting teaching and learning (e.g. Penn MOO; Diversity MOO; see also discussions by Hand and Skipper, 1996; Bruckman, 1998; O’Day et al., 1998), for building social and vocational communities (e.g. Curtis, 1996; Mynatt et al., 1998; Schank et al., 1999), for virtual conferencing (e.g. Towell and Towell, 1997), for exploring issues of culture (Chapter 13) and for supporting collaborative work activities (e.g. Doppke at al., 1998; Evard, 1993; Churchill and Bly 1999a,b,c; Tomek and Giles, 1999).

Increasingly, many of us find our working life having to work not only in a main office, but also at home, at other institutions, from hotel rooms and so on. Such multiple sites, however, create problems in that all too often we find our documents and our colleagues unavailable at our current physical location. In a world where telecommuting is the norm, it seems plausible that a stable shared virtual office might be of benefit, creating a set of places that can be accessed to locate content and meet colleagues. This chapter will describe the issues that have arisen from three years of regular use of collaborative 3D virtual space built within AlphaWorld, a popular Internet-based shared 3D world. A small virtual “office” was constructed to try to understand the possibilities and the problems of this technology in actual use. It is these possibilities and concerns that will be discussed.