Covering The World

BACK IN 1990 MICRO-soft overlord Bill Gates and cellular-telephone pioneer Craig McCaw introduced what was arguably the most audacious plan in modern telecommunications: a network of satellites--hundreds of them--forming a worldwide "Internet in the sky," beaming data anywhere in the world 35 times faster than the fastest analog modem. This wasn't just a couple of Seattle guys batting around an idea over lattes; they formed a company, Teledesic, to launch more than 840 satellites in a decade and create a global information infrastructure from scratch. Analysts blinked a few times at the info titans' scheme, and then said it couldn't be done.

Last week that assessment shifted a bit. Another member of the Seattle industrial pantheon joined Teledesic: Phil Condit, head of the Boeing Co., threw in $100 million to make his aerospace firm an equity partner. At the same time Boeing became Teledesic's prime contractor, getting roughly $9 billion to build and launch a scaled-down array of 288 satellites, plus spares, over the next five years. If it works, Gates, McCaw and Boeing are going to get even richer, and the world is going to get a lot smaller. But a few obstacles still stand in the way.

Of the more than a dozen satellite systems that plan to provide high-speed multimedia access, three--SkyBridge, M-Star and Teledesic--will rely on a network of Low Earth Orbit satellites (LEOs), just a few hundred miles up. Satellites like the ones that beam down television signals today are in geosynchronous orbit 22,000 miles above fixed points on the equator. From the lower altitude it takes less power to get a signal back to Earth, which means smaller antennas. And because LEOs are closer, there's virtually no delay in the signals they send back--the "latency" inherent in transmissions relayed by geosynchronous satellites. But LEOs also "see" less of the planet's surface at a given time, so the birds must hand off signals like cellular-phone base stations do. In space, that's a complicated bit of prestidigitation.

What Teledesic gets for its trouble is the ability to sell anyone, anywhere, access to hunks of data with a dinner-plate-size antenna that today would require a high-powered connection called an E1--two megabits per second, compared with a modem's typical 28.8 kilobits per second. That's enough bandwidth for a corporation to do transnational videoconferencing. A developing nation could use Teledesic to leapfrog copper cable. "Boeing's leadership on Teledesic's industrial team will help us play a role in bringing the advantages of the Information Age to the entire world," McCaw said in a press release, "particularly to those being left out as a matter of cost or geography."

Still, no one has ever completed a constellation of LEOs. It's not easy; roughly one out of every 30 satellites launched never makes it to space, and satellites are still, to some extent, hand-built. Iridium, a company planning to provide worldwide mobile-phone service, has been postponing the launch of its first five LEO satellites since January; they hope to launch this week. Naturally, Teledesic representatives express only confidence. "There's nothing there that hasn't been done before," says Teledesic CEO David Twyver. "If you get the technique right, there's no reason you can't scale it up." Boeing will have to learn to build satellites quickly and cheaply, but expanding into satellite communications could buffer it against the financial ups and downs of the aircraft industry.

Even if it all works, will anyone pay for it? "There isn't really a market plan that has convinced me that any of these guys can be successful," says Phillip Redman, a senior analyst with the Yankee Group. Teledesic's Twyver says corporate demand for low-latency bandwidth is going to explode over the next five years. But the Yankee Group forecasts only 60 million subscribers worldwide for satellite systems by the year 2010, while plain old cell-phone numbers are expected to hit half a billion. "The people in first class might pay a couple thousand dollars for a phone that's going to cost $3 a minute to talk on," says John Pike, the director of space policy at the Federation of American Scientists, "but my hunch is the rest of us in the back of the plane are not." In fact, no one knows whether any of the plans will fly: of the three systems using broadband LEOs, only Teledesic has FCC approval. But when you combine audacity, cash and technology, the sky is the limit.