The technicians at BT's National Control Centre in Oswestry, Shropshire, knew nothing of the promotion that Tuesday night until the map on their giant video screen began pulsating with red light over Bristol. Some 200,000 callers tried to reach that lone number in the first 15 minutes. Only 50 got through, but the rest tied up every line in the Bristol exchange. If anyone in the city had tried to dial 999, they would have found their lifeline blocked.
By the end of that frantic quarter-hour, the technicians had solved the problem, identifying the target number and reprogramming exchanges across the country to filter out calls to it. Typically, such filters check to see whether they have put through a call to the number in the previous two minutes. If they have, they return a busy signal.
Traffic jams like the one that hit Bristol are relatively easy to clear using today's technology. When France's system became clogged during the 1992 Barcelona Olympics, calls were automatically re-routed through other countries, says Graham Whitehead, BT's advanced concepts manager.
But the globalnetwork now faces a more consistent threat. Against a background of steadily increasing demand for voice telephony has arisen the Internet, with its ravenous appetite for bandwidth and disregard for distance.
While technology should be able to cope with the dramatic increase in use over the next couple of decades, commercial and economic factors, combined with the lead time needed to install new lines, could yet leave a gaping hole between demand and supply. "Data networks will break down because the pricing model won't support increases in demand," says Adrian Kamellard, the corporate futurologist at Cable & Wireless.
Demand for telephones has been growing ever since 1876, when Sir Alexander Graham Bell first said: "Mr Watson, come here, I want to see you." International telecoms traffic alone more than trebled from about 16 billion minutes a year to 53 billion minutes in the decade to 1994, according to Cable & Wireless. The first transatlantic fibre-optic cable, TAT8, laid in 1988, was full within six months. Three more have been built since.
"There's now a glass belt around the world," says Mr Whitehead.
The Earth's new girdle is needed partly because of rising standards of living in developing countries. The globalisation of business also requires more overseas calls, and the ease with which people can move around the world means more tourists and expatriates phoning home.
The domestic picture varies from country to country, but even in the developed world demand is rising. A typical household which used to have a single line into the home plus, possibly, one in the breadwinner's workplace, could soon support close to a dozen connections; three each in the parents' offices - for phone, fax and Internet - a mobile in the car, another in a handbag and, at home, extra lines for the kids and the computer.
In the future people will want greater access, not just in terms of the number of connection points, but in the length of time they are on line and the volume of information they want to send and receive.
Miniature headsets that transmit sounds through the cheekbone could give people constant contact with the Internet's "talking" successor - call it the "datasphere" - as well as guaranteed voice telephony links whether the wearer is at Bank station or on the ski slopes of Mont Blanc. And most everyday electronic devices will be linked in some way to the global telecoms net.
"For most people it will be close to the computer model," says Mr Kamellard. "Ninety per cent of the population use computers now but don't even realise it." Other uses are already boosting demand. Staff at Analysys, a telecommunications strategy consultancy, use videophones in their office.
Andrew Entwistle, the principal consultant, says: "It mirrors face- to-face communication. Nobody has ever yelled at me on the videophone."
The price for this new technology, however, is a high-capacity ISDN line, which typically costs pounds 400 to install plus pounds 80 a month to maintain.
The biggest single factordriving telecoms growth in the developed world is the Internet. The number of subscribers is estimated to double every year, while their use of the system doubles every six months.
In the past year, data transmissions over phone lines have for the first time eclipsed voice. In California there have been cases where urgent calls could not be placed because Internet users had tied up all the local circuits. A reversal of the Bristol scenario, but just as dangerous for anyone ringing for an ambulance.
If the potential in sophisticated markets sounds impressive, try adding on the future growth possibilities in less developed countries. Almost three quarters of the world's 5 billion people are not hooked up. "There are more telephones inside the M25 than in all of Africa," says Mr Whitehead. China, with its billion-plus people, could equal the West for telecoms demand if its economic growth persists over the next 20 years.
This growth is not likely to put pressure on the backbone links between exchanges. For the most part these have already been converted to high- capacity fibre-optic cables. Overall, the telecoms industry has a massive surfeit of capacity. "Almost all of the telephone network is sitting doing nothing almost all of the time," says Mr Entwistle. With the right hardware at its ends, a fibre-optic cable could handle 5 billion conversations - "enough to allow every man, woman and child on the planet to talk at the same time, twice over," says Mr Whitehead.
But the hardware at the ends of the cables, and the links from there to the home, are potential bottlenecks. Exchanges, or switches, are specialised computers, with price tags comparable to mainframes of the 1970s. Upgrading costs a fortune.
The same is true of the copper lines that lead from the local exchange to the home. Copper wires have a miserably low bandwidth - a measure of the amount of data that can pass in any given period of time. As people use their communications links to carry more information, they will become increasingly inadequate.
Broader pipelines exist, from ISDN lines through co-axial cable -the kind used to transmit television - to fibre-optic cables. The price per foot of these lines is low, but the labour-intensive job of laying them in the ground is not. And the cost will be partially borne by the customer, who will have to pay for modems, data cards, set-top-boxes or, possibly, radio dishes.
Initially the cost per customer of switching to a higher standard will be exorbitant. After being burned by buying technology that six months later halved in price, many will be reluctant to be first. Only if business leads the way, or some entrepreneur takes a gamble, will new technology sweep through the economy.
"I hate to say it but we need more men like Rupert Murdoch who are willing to gamble on buying a million units of a new technology, selling them at a subsidised price and then getting it out of the customer later," says Mr Entwistle.
The telecoms companies may well take on that role, but at the moment their enthusiasm is waning, and will not revive until the way prices are set is modified. "The telcos are investing now, but none of them could draw up a P/L statement showing they're making a profit out of it," says Mr Kamellard.
In North America, local phone calls are free, or rather, they are paid for by a flat fee. In the early days of telephony the strategy helped to boost usage, and even today Americans use their phones four times as much as Britons. That was fine when the average conversation lasted three minutes, but the Internet has boosted that figure to close to 30.
While the US's exchanges desperately need upgrading to meet this demand, the local phone companies will get no return from their new investment. To solve the problem they are lobbying to have Internet service providers - the first contact point for cyber-surfers - to pay access charges just like long-distance companies. These costs would be passed on by the providers to their users.
Britain faces the opposite problem. BT charges by the second, thus ensuring that it has a rising cash flow to meet rising demand for new capacity. But 12 years after privatisation, its pricing strategy is still based on a monopolistic model, says Mr Entwistle. It is prepared to lose market share while pricing its product to maximise profits. "It's not wicked. It's completely standard, rational behaviour."
Pricing calls to use up any excess capacity does not make sense for a company with the market dominance of BT. Because the market is not perfectly elastic, a 10 per cent cut in prices generates only a 7 or 8 per cent increase in volume, so the company would lose money. Since 1985, BT's price per minute for national calls has more than halved from 7p to 3p, according to Analysys. But the drop has only come in response to its competitors' cuts. And it pales by comparison with the real marginal cost of a call, which is only fractionally above zero.
The same logic argues that BT should ignore rising demand as long as possible. Only when congestion gets so bad that deserting customers force down its profits will it be worth while investing in new kit. Increased competition, only now starting to bite, will be the best cure for this ill in the medium term, forcing BT to concentrate on maximising volume instead of margins.
In the long run, competitive pressures on either side of the Atlantic - and in any other deregulated markets - will ensure plenty of capacity to go around, at ridiculously low prices. Mr Entwistle sees a day when it will be economic for some people to buy flat-rate packages allowing them to call anywhere in the world, for as long as they want, for no extra charge. "It will be the death not only of distance," he says, "but of time."Reuse content