AT a recent conference my vision of a future world dominated by accelerating change and concentrated waves of demand leading to chaos, was publicly challenged. The questioner proposed that while this might be true for a world of bits, a world of atoms would turn out to be stable and far slower moving. He cited Low Earth Orbit satellite systems as the future route to low-cost access and high bandwidth to the home and office. And thereby our problems were solved forever and we could expect to see network stability.

I suspect he had gained this view from the popular and trade press fed by financial markets desperately trying to pick the winners. The reality is that satellite systems of any complexion will never provide such capability. Radio bandwidth is fundamentally limited and satellite systems will always be expensive to engineer, launch and maintain. Today satellites provide less than five per cent of our international circuits, and well under one per cent of the total intra-country capacity.

It costs £10 billion to assemble a workable number of orbits for global coverage, and the bit rate is limited to a few thousand bits a second. Even with a simultaneous bandwidth increase and thousandfold reduction in costs LEO satellites would not compete against fibre or copper cables. Their key attribute is wirelessness - mobile and portable access anywhere, unless that is, you are shielded by buildings or mountains.

So, are they safe in the mobile arena? Not especially, there is competition. Second and third-generation terrestrial mobile systems offer more in all respects. And so will airships or balloons hovering at 25km, and pilotless drone aircraft flying in circles above cities at 15km.

At a height of 25km the atmosphere is free of weather and aircraft, and balloon technology to hover above a city is cheap. At a height of 15km a drone aircraft can circle for more than three months without refuelling.

It is interesting to reflect that at the start of the LEO development, neither drones nor balloons were on the technological horizon. But such is the pace of technology that before the first LEO satellite was launched it was evident that they would soon be under threat.

If I am out in the bush of Australia, up a mountain in America, or on a boat in the middle of the Pacific, I will look to a LEO system for instant access and communications. But in my home, office, and travelling most of the world, I think not: there are better and cheaper options. I can no longer think of an area or sector of technology where this situation does not obtain. All technologies can and will be eclipsed, and for most their demise will be fast. Only a few seminal aspects will enjoy longevity, most will be overtaken before they even get to market.

The seminal discovery powering the information technology world is the electromagnetic wave, supported by the fundamental properties of the materials employed. Even our key hardware, which at first glance appears stable, is evolving generation to generation. Optical fibre, radio, transistors, integrated circuits, lasers and other components are achieving performance levels unthinkable a decade ago. The only exception seems to be our system software, which, for the most part, seems to progress in the opposite direction.

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