No problem over property at the centre of vision

As the UK's media providers prepare to go digital, Peter Gibbins introduces a unique research company in more ways than one - the virtual centre of excellence in multimedia and digital broadcasting

Nicholas Negroponte, director of the celebrated MIT Media Lab, remarked in Wired magazine a couple of years ago that a laboratory is like a dog, only worse. "While dogs age at a factor of seven," he wrote,"labs age at a factor of ten . . . so how does a lab avoid sclerosis?" His answer is to employ bright students: "I run a company with 300 employees and 20 per cent turnover each year . . . The incoming lot are always between 16 and 25 years old . . . That 20 per cent is the fountain of youth."

Industrial research laboratories lack that rejuvenating current. I have noticed that after an initial burst of enthusiasm, industrial labs often mature into an early middle age five years or so after set up. Commercial pressures tend to push research towards the short term, then towards product development, and finally engineering.

Industry therefore collaborates with academia not out of the purest altruism, but out of a shrewd self-interest. As well as access to the energetic young, pre-competitive collaborative research can usefully impact on standards. It can also be cheap and it can create a space in which longer-term research can flourish. In the booming fields of digital broadcasting and multimedia these virtues lie behind the Virtual Centre of Excellence in Digital Broadcasting and Multimedia Technology Ltd, a consortium of industrial members and universities which trades under the more manageable name of the Digital VCE.

Broadcasting and network multimedia are especially ripe for collaborative research. Both depend entirely on standards. New products and services will depend on technology yet to be invented, precompetitive research yet to be done. During 1998 the bemused consumer will be hit with a number of goodies from companies and joint ventures most often called "B Something B". Suddenly many channels of digital television will be available in the United Kingdom: via satellite from BSkyB, via cable from cable companies, and via regular terrestrial television aerials from British Digital Broadcasting (BDB).

Digital television with a modest degree of interactivity will be delivered by British Interactive Broadcasting (BIB), a consortium of BT, BSkyB, Matsu****a and Midland Bank. Watching a Formula One race you may, for example, view the event from the seat of Ralph Schumacher, Damon Hill or your driver of choice.

Broadcast data is going digital, and so TV and the PC are converging; or, as BT says, "colliding", it is happening so fast. A small computer called a set-top box (STB) will turn streams of bits received via satellite, cable or the ether into pictures and sound, or whatever. Everyone agrees that we are only at the beginning of this revolution in entertainment, and then in work and the way we structure our lives. New, as yet unimagined products and services demand huge new efforts in R&D.

UK universities already do first-rate work in digital broadcasting and multimedia, but the research is widely dispersed. There is no centre of excellence in the UK to match, say, the MIT Media Lab. The DTI's Technology Foresight panel report on communications published in 1995 recommended setting up several "virtual centres of excellence", one in digital broadcasting and multimedia. With commendable urgency, the Digital VCE was incorporated last April. (Another VCE, in mobile and personal communications, had been established late last year.) The Digital VCE is an unusual company, limited by guarantee, managed by a board of directors, but without shareholder capital. The industrial members pay an annual subscription and the VCE bids for additional funding to support the research it does. Its central mission is to do basic research and generate intellectual property in its eponymous technologies.

Everyone who has been involved in collaborative R&D knows that the issue of intellectual property rights is a thorny one. Companies, and even universities and academics, are sensitive to giving it away and with good reason. We live in a knowledge-based, information society. More and more of the value of goods and services is bound up with intellectual property. A key patent can be worth a large fortune.

The VCE's intellectual property regime succeeds in being preternaturally generous to all sides. The VCE, as a company, will own the intellectual property generated. On the other hand all the members, industrial and academic, will have full, worldwide, non-exclusive, royalty-free access to it. So will the industrial members' parent companies and subsidiaries, an arrangement which is very attractive to inward investors. Industrial members can also licence the intellectual property to their commercial partners when necessary. As far as the member universities are concerned, all are free to use all the intellectual property in research.The particular university that generates the intellectual property can license it outside the VCE commercially.

The VCE is based in Oxford Science Park. Its academic members are the universities of Bradford, Bristol, Cambridge (in association with Imperial College, London and the National Film and Television School), Essex, Surrey, and University College, London. They were selected in 1996 by the then prospective industrial members of the VCE . On Monday, the VCE will be formally launched at the Department of Trade and Industry by Barbara Roche, parliamentary under-secretary of state for small firms, trade and industry. Its initial projects, coordinated by Bristol and Surrey, will have just begun, as will what I take to be the VCE's biggest challenges: doing the best possible research, maintaining the commitment of the industrial members, and doing real technology transfer, attracting the largest possible industrial membership. We should know within three years whether a VCE, with the support it can command from industry, academia and the state, is an effective vehicle for delivering these results.

Peter Gibbins is executive director of the Digital VCE. He is also visiting professor in computer science at the University of Exeter.