Getting your Trinity Audio player ready...

Everyone wants fiber-optic speeds in their data connections, and lots of companies have been planting vast amounts of fiber to carry so-called Internet backbone traffic. Everyone also tends to agree that optical- fiber wires to homes and businesses would be a great way to bring high-bandwidth–that is, fast–Internet connections.

But the cost of laying all that fiber would be immense.

Now, a Seattle company says it has found a way to bring fiber speeds without the wires–using light waves to carry data.

TeraBeam Networks, which is privately held, has been working in stealth mode since 1997. The company got its first major public attention earlier this month, when it announced it had lured Daniel R. Hesse, president of AT&T Corp.’s Wireless Group, to become TeraBeam’s chief executive. The technology itself was unveiled last week at PC Forum.

While the news of a big old economy deal, Tribune Co.’s planned purchase of Times Mirror Co., was greeted with profound and almost universal indifference, TeraBeam’s presentation created a significant stir. In a telling commentary, I listened as an investment banker asked Greg Amadon, TeraBeam’s founder, chairman and chief technology officer, if it was too late to get in on the company’s next round of funding.

If TeraBeam can deploy this bandwidth as it claims, it has a chance to own what people sometimes call a disruptive technology–a breakthrough that wrecks old business models and inspires new industries.

The company is aiming what it calls its “fiberless optical” networks at metropolitan areas, with transceivers on customers’ premises exchanging data with TeraBeam devices installed around a city in a pattern somewhat like a mobile phone company’s overlapping cells.

But this isn’t about radio waves. Light itself will carry the signals.

The advantages, if this works as advertised, are profound. First, the bandwidth will be enormous–several gigabits per second. A gigabit is 1,000 megabits, and a megabit per second is more than 20 times as fast as a home modem connection.

Second, because it’s wireless, the company won’t have to negotiate for rights-of-way and dig up streets.

Third, a customer won’t need to mount anything on the roof; the transceiver, about the size of a satellite dish, can send and receive data through a window.

The TeraBeam service is aimed at businesses, universities and other large enterprises that want to get big data bandwidth at what could be a very low relative cost. Still, the relentless progress of technology suggests this kind of thing might well be available to consumers in not too many years.

TeraBeam says the service will be deployed commercially this summer. It aims to cover the top 50 U.S. cities and some international cities within three years.

So, what’s the catch? For one thing, the technology requires clear line of sight between each data node. It doesn’t work as well in dense fog, which is why TeraBeam’s test bed has been misty Seattle, Hesse said. More cells would solve that problem, he added.

For another, it’ll take real money to deploy this, even if it does turn out to be vastly cheaper than the competition. The company has already raised $21 million in seed and early venture financing rounds, and is just about to close a third round of about $100 million, Amadon said.