This page still under construction.
Introduction
1.1. What is a Collaborative Virtual Environment?
Collaborative Virtual Environments (CVEs) involve the use of distributed virtual reality technology to support group work. A necessary, but not sufficient, condition for a CVE is the provision of simultaneous multi-user access to a virtual reality system. However, there may be a world of difference between a multi-user system and one that actually support cooperative work and so a second condition is that the system must explicitly consider and support the needs of users who wish to work together,
The essence of CVEs is that users are explicitly represented to each other within a shared space. Furthermore, they should be free to move around within this space, encountering each other and also objects and information of common interest. The interactive nature of true virtual reality systems means that they should also be able to interact with each other and with the objects and information.
1.2. Why develop Collaborative Virtual Environments?
There are several general reasons for supposing that CVEs might be useful. Three general ones are:
1. Support for natural spatial social skills - as inherently spatial creatures, human beings possess powerful spatial skills. The development of single user interface technologies has involved exploiting an individual's cognitive spatial skills - i.e. their ability to reason about space (e.g. spatial classification and navigation abilities). However, humans also posses social spatial skills whereby they make use of shared space as a means of negotiating interaction with one another. The importance of eye-contact in conversation, mingling at cocktail parties and behaviours such as queuing, jostling and scrumming are examples of how such skills are used. All of these are highly dynamic, subtle and adaptive behaviours and are a far cry from the kinds of "floor control" mechanisms that have previously been proposed for communication systems.
2. Inherent scaleability - nearly all real-time CSCW systems (e.g. multi-media conferencing and media spaces) have focused on small scale interaction between a few concurrent users. There is a real need to address synchronous communication between many hundreds of thousands of simultaneous users. In order to do this, the underlying approach must have some inherent property that is scaleable. Shared space is one such property as it provides a natural an intuitive way to interact with many people and yet separate what is immediate and local from what is peripheral and distant. Thus, one can talk to a few people nearby and yet see potentially thousands of others. Those who are sceptical about the possibilities or need for large scale interaction might like to spend a little time ruminating on the possibilities for communication, awareness and interaction within, say, a large sports stadium or at a large public performance (which may peak at around 150,000 simultaneous users).
3. Applicability to cooperative spatial tasks - some highly spatial tasks such as 3-D design (e.g. architectural design) and environmental planning which are currently the domain of single user virtual reality applications may also offer possibilities for collaboration.
These three general motivations underpin much of our work and we will often return to them in the following chapters.
1.3. State of the art
The emergence of CVEs can be related to, and has been influenced by, several areas of technology.
First, a number of projects have been explicitly addressing the issue of co-operative work and virtual reality. These include the COMIC ESPRIT III Basic Research Action [Benford93] and the UK's Virtuosi project funded under the EPSRC/DTI CSCW Programme [Benford94b]. The former has been developing general techniques for CVEs, many of which are covered by this tutorial, whereas the latter has been focusing on the industrial application of these and other techniques through two major pilots; the virtual factory for the manufacturing industry and the virtual catwalk for the fashion industry. Japanese efforts include Collaborative Workspace from NTT [Takemura92], an immersive system supporting real-time facial action capture and animation and also NTT's Interspace project which supports groups of people navigating a virtual city in order to go shopping and engage in recreational activities [Suzuki95]. The Greenspace project should also be noted as an example of a demonstration of a cooperative virtual space between the HITLab in the US and sites in Japan.
Probably the most influential project in this area has been DIVE, a development at the Swedish Institute of Computer Science (SICS) which has resulted in a freely available CVE which has been used world-wide as the basis for a host of other projects [FahlŽn93, Carlsson93]. DIVE provides a general development environment for CVEs and more details can be obtained from Lennart FahlŽn (lef@sics.se).
Within the specialised domain of military simulation, the US Navy has developed NPSNET (based on the earlier SIMNET) which allows groups of several hundred people to engage in networked VR battle simulations [Macedonia94].
A number of other research projects and systems have also tackled distributed systems and network issues surrounding networked virtual reality (although with less emphasis on co-operation). These include the MR Toolkit [Shaw93] and Aviary from Manchester University [Snowdon94].
Most commercial VR systems also provide some limited support for networking including dVS from Division Ltd. (UK), Elysium from Virtuality (UK), Superscape from Dimension International (UK) and the World Toolkit from Sense 8 (US).
Outside the domain of virtual reality, three other areas of research should be mentioned.
First, there have been a number of projects considering the use of 3-D meeting room metaphors for multi-media conferencing systems including Milan [Leevers92] and the Electronic Office of Cook [Cook91].
Second, research into Media Space [Gaver92], a major topic in its own right, has emphasised issues of awareness, embodiment and privacy that have influenced initial thinking in CVEs.
Third, the development of Multi-User Dungeons (MUDS) has given insights into largish scale real-time interaction in text-based environments [Bartle].
1.4. Structure of this tutorial
This tutorial provides an overview of CVEs, divided into three main issues:
the people - issues of representing people to one another;
the place - issues of structuring shared spaces and information visualisation;
the activity - how to support communication between the people within the place.
1.5. Acknowledgements
We would like to thank the following for their support in funding this research: EPSRC though the Virtuosi project and PhD studentships. The European Commission through the COMIC ESPRIT III Basic Research Action. We would also like to thank the following researchers who have contributed to this research: John Bowers, The University of Manchester; Lennart FahlŽn, The Swedish Institute of Computer Science; Tom Rodden and John Mariani, Lancaster University; and Research staff at British Telecom, Division, BICC, GPT, GEC Hirst Research Centre, Nottinghamshire County Council and Nottingham Trent University for their contribution to the Virtuosi Project.