In telecommunications we currently use copper and fibre cables, free space optics, terrestrial and satellite radio as the traffic lanes for bits. The way we organise information on these bearers relies on a limited range of opportunity. Time, space, frequency, wavelength and code are the principal degrees of freedom.

In broadcast systems we use frequency or wavelength (they are the same thing expressed in two different ways) slots to carry and identify individual radio and TV stations. When digital broadcast takes off we will also include time, and, later, coding as additional channel and information selection mechanisms. In broadcast no attempt is made to use space as a selector - information is sprayed everywhere - everyone can see it. The same is generally true of satellite and mobile telephones, where the use of space as a selection mechanism is minimal.

In contrast, free space optical systems and microwave radio links employ spatial focusing to constrain signal energy to defined physical locations. Alternatively, copper and optical fibre cables route information with great precision straight to the office and home, telephone, television, computer or whatever.

So when we examine telecommunication and computer networks we find a very interesting range of selection processes. In telephone systems time and space are dominant, and this is also the case for local area networks. Cable TV on the other hand is dominantly analogue and employs frequency and space. Few, if any, network systems employ all the degrees of freedom - time, frequency/wavelength, space and coding - for effective communication. Most use two.

Another important network feature is the information flow symmetry, or lack of it. This spans broadcast and CATV that are totally asymmetric (studio to listener and viewer), LANs (often partly asymmetric, depending on the user group) and the telephone which tends to be wholly symmetric. What a soup of interesting combinations our world has become.

From all the media hype and political hyperbole we might be excused for thinking that someone somewhere had a grand plan, a design, for the information superhighway. Unfortunately this is not the case. No one knows how to combine all the options to deliver broadcast radio and TV, Internet, telephony and more simultaneously in a unified way. It seems clear that we have to combine all the modes - time, frequency/wavelength, space and perhaps coding. Crudely, what a lesser choice gives us is: the telephone network - a footpath; cable and broadcast - a one way street; LANs - a road under constant repair; cellular radio - a country lane. Unless we simultaneously access time, space and frequency/wavelength, we stand as much chance of realising a superhighway as getting the Queen Mary cruising on the M25 - and the delays will be similar.

Peter Cochrane holds the Collier Chair for the Public Understanding of Science & Technology at the University of Bristol. His home page is:
http://cochrane.org.uk