(5 of 7)
The problem comes when you get off the turnpike onto the roadways owned by local phone companies and cable-TV operators. Some of these are being converted to high-bandwidth fiber optic. But at the end of almost every local system — the "last mile" that goes from the local-service provider to the house — you run into the electronic equivalent of a bumpy country road. In the phone system, the bottleneck is that last bit of copper wiring, which seems far too narrow to admit the profusion of TV signals poised to flow through it. In cable TV, the roadblocks are the long cascades of amplifiers that run from the company's transmission headquarters to the home, boosting the signal every quarter-mile or so. These amplifiers are notoriously unreliable and generate so much electronic noise that two-way traffic in a cable-TV system is all but impossible.
It has long been assumed that nothing was going to change much in telecommunications or television until fiber was brought all the way to the home, a Herculean task that was expected to cost $200 billion to $400 billion and take more than 20 years to complete. The breakthrough that is bringing the info highway home much sooner than expected is the discovery, by both the phone companies and the cable industry, that it is possible to get around the bottlenecks in their respective last miles without replacing the entire system.
For the cable-TV companies, the key insight came in the fall of 1987, when $ cable engineers demonstrated that coaxial wire could carry information quite effectively over short distances; in fact, for a quarter-mile or so, it has almost as much bandwidth as fiber. They pointed out that by using fiber to bring the signal to within a few blocks of each home and coaxial cable to carry it the rest of the way, the cable companies could get a "twofer": they could throw away those cranky amplifiers (giving them a system that has more capacity and is easier to maintain) and get two-way interactivity almost cost- free.
For the phone companies, the breakthrough came three years ago when scientists at Bellcore, the research arm of the Baby Bells, found a way to do what everybody had assumed was impossible: squeeze a video signal through a telephone wire. The technology, known as asymmetric digital subscriber line, has some drawbacks. It cannot handle live transmissions, and the picture it produces is not as clear as that provided by a well-tuned cable hookup — never mind the high-definition TV signals expected to come on line before the end of the decade. Bellcore researchers say they have already improved the quality of the picture and that with further compression they may be able to accommodate several channels of live video.