Special Report on Multimedia: Another world is virtually here: A fast-developing technology is now opening doorways to our dreams, writes Tony Feldman

ALTHOUGH few people are sure exactly what it means, virtual reality is a term widely used in a wealth of popular and professional contexts.

The uncertainty about its meaning has fuelled plenty of speculation about VR and what it can do. In the popular imagination it has grabbed attention as a technology which offers a doorway to our dreams. Everything we have ever wanted to do is apparently possible. All you have to do is don a VR helmet, slip on a data glove - or a full-body suit - and VR will supply the experience without physical commitment, inconvenience and danger of the real thing.

This vision may well come true. Indeed, some serious VR specialists are convinced that within the next 50 years we will be able to participate in totally convincing artificial experiences of almost any kind. The social and ethical issues raised are almost as complex as building the technology to make it possible. For the foreseeable future, therefore, it is probably best to stick to considering the impact of applications which are less sensational but more likely to be easily within our grasp.

Although definitions vary, most professionals agree that virtual realities are computer-generated worlds which can be explored in real time.

The idea of 'real time' is crucial. In simple terms, it means that the world is continuously re-computed as it is explored. A virtual world is not a pre-recorded experience stored on a magnetic or optical disc waiting to be accessed. It is created 'on the fly' by the computer in response to whatever you do inside the world.

Imagine having your senses mapped into a computer-generated world so that you seem to be walking down a virtual street. As you turn to look left, the landscape swings across your vision just as if it were real. The thousands of computations needed to create the seamless change of perspective are performed by the computer at the same time as you carry out the movement.

This is the ultimate form of interactivity because within such real time environments, the only limits on where you can go or what you can see and do are those imposed by the rules in the underlying computer program which effectively provides the 'natural laws' governing the virtual world. If the laws forbid you from flying, you cannot fly. If they prevent you from walking through obstructions, you will bump into walls rather than float through them.

The real-time quality of VR needs vast computing power if it is to produce anything even remotely convincing. To generate a worldscape which looks like the real thing requires extremely sophisticated software which not only models geometrical shapes from every possible perspective, but also renders surface textures and lighting effects accurately.

Despite the popular hype, virtual reality is largely an industry struggling to get out of the research laboratory. Although the picture is changing and the first genuine commercial applications are now emerging, the technology still lags behind the requirements of people who would like to put VR to work.

There is plenty of talk about applications and even a good number of pilot projects. In Britain, for example, Glaxo, IBM, the University of York and the VR specialists Division have joined forces to pioneer a VR molecular visualiser. The Leicester-based VR pioneer, W Industries, is selling VR computer games. London University's Parallel Applications Centre is working with Thorn-EMI to create an architectural design system. The University of Nottingham is experimenting with VR for teaching children with special learning needs and British Aerospace is studying the use of VR in displays for fighter aircraft.

There is real progress but the pace is much slower than the VR enthusiasts would like.

Where do VR and multimedia meet? Many in the VR industry would like to forge links if only because multimedia appears to have an immediate market for its products and the VR specialists would like find a short-cut to commercialising their own technology. The multimedia industry is attracted to VR because it is anxious about the pace of market development and the timescale for paybacks on its massive investments.

Beyond this, VR is an intriguing technology based on the manipulation of sophisticated multimedia environments. So it is likely that VR should be able to enrich and extend multimedia applications.

Attractive as the idea may seem, however, VR and multimedia do not fit easily together. The key reason is that VR is essentially a real-time phenomenon while most multimedia is not. The only significant exception is in the field of video conferencing, where multimedia and the telepresence aspects of VR pass very close.

More generally, however, multimedia environments tend to be based on stored data such as digitised images, sound and text, accessed under relatively simple software control according to a user's responses. This is light years away from virtual worlds, generated and regenerated by state-of-the-art 3D graphics software running on fast, powerful computers.

The only practical link so far suggested is to use VR as an intuitive interface for users of complex multimedia databases. The idea is that VR might provide a familiar and simple metaphor for finding what you want in a massive storehouse of multimedia information. A suitable VR interface, for example, could give the user the sense of being in a real library. Finding and retrieving information might then be reduced to using virtual index cards drawn from virtual filing cabinets. While the underlying metaphor renders the process of search and retrieval friendly and easy, the effects of using the facilities of the virtual library could be made much more powerful than those in a real one.

It is significant that so far the only link suggested is one that reduces VR to a tool that makes it easier to gain access to multimedia information. This is like using a sledge-hammer to crack a nut; why go to the trouble of immersing people in expensive virtual worlds just to help them get at multimedia? Why not make multimedia friendly and accessible in the first place?

Tony Feldman is international editor of the newsletter Virtual Reality Report and chairman of the UK's annual virtual reality conference.

(Photograph omitted)