By Leslie Walker
Washington Post Staff Writer
Thursday, October 7, 1999; Page E1
And now, from some of the same people who brought you point-and-click images on the World Wide Web: an encore.
The Geocast Network is being designed by members of the team that created the original Netscape browser. Under wraps for two years while its inventors developed a prototype and raised money from three of Silicon Valley's leading venture-capital firms, Geocast is one of the more innovative plans I've seen for marrying broadcast television with the Web.
The prototype I glimpsed this week in a hotel suite in Washington convinced me that the Internet still has a baby face with many features yet to be fleshed out that could make it almost unrecognizable by the time it grows up.
So far, all bets on the "convergence" of television and computing have been that it will happen one of two ways: either by adding Internet connections to your TV through a set-top box, or by embedding a standard television receiver into your personal computer. Top players in the television, telephone and Internet industries are racing to implement variations on those plans.
Enter Geocast Network Systems Inc. of Menlo Park, Calif., a hybrid whose next-generation idea is to use digital-television signals to zap full-motion video to people's computers at high speeds, then tap their slower Internet access accounts to carry data back to Geocast. The goal is to create a new communication channel that would combine the talk-back ability of the Internet with the one-way "multi-casting" power of television.
The service would be free to users supported by a mix of advertising and commerce revenue but would require people to purchase a receiving device, for less than $300. The device includes an antenna and a disk drive to allow storage of video and data beamed to PC users during off-peak hours.
Geocast is one of the first plans for integrating digital TV signals with the Internet through special browsing software. Since others are being worked on feverishly behind closed doors, it offers a tantalizing hint of what's to come.
The significance of the idea lies largely in the difference between the airwaves of television and point-to-point connections of the Internet. Television sends signals from one point to many simultaneously, as opposed to the Internet's "uni-casting" way of transmitting from one single point to another. The Net's architecture is why it currently takes so long to send full-motion video: The more people who want to access any Internet file, the more network resources are required. Broadcast TV signals, on the other hand, can be received by millions of people at once without affecting transmission quality.
Geocast wants to use the strengths of each medium to overcome the weaknesses of the other. It is made possible partly by the pending conversion of the nation's television broadcasting system to an all-digital format, which is more compatible with the digital DNA of the Internet. The federal government has mandated that all 1,600 U.S. stations switch to digital by 2006.
Geocast's founders plan to send their data over the portion of the digital TV spectrum that will remain unused even as the TV stations broadcast their own super-high-quality digital images. They say their hybrid system will deliver data to computers much faster than the other high-speed interactive systems under development, including enhanced cable and telephone lines.
"I think this is one of the better ideas I've heard about to tie in use of the digital spectrum with broadcasting and the Internet," said Chuck Sherman, executive vice president of the National Association of Broadcasters.
In addition to its unusual technology, Sherman said, Geocast's business strategy of forming a national network by giving affiliated local TV stations a slice of advertising revenue and letting them share in national programming could prove attractive to stations because it is so familiar. Of course, the business won't get far unless many stations do sign up, because its economics require national reach.
Today, Geocast Network Services Inc. takes a step toward that goal when it announces that it has signed a deal with the Hearst-Argyle Television Inc. group of 26 television stations. The deal licenses a portion of Hearst-Argyle's digital broadcast TV spectrum for Geocast to launch its service late next year. Geocast is negotiating with other station groups for similar deals.
Hearst is investing up to $10 million for an equity stake in Geocast and access to a network it hopes will offer a new way to sell local advertising and deliver more personalized information. "This allows us to use both the mass reach of broadcast television and the one-to-one capability of the Internet to develop a deeper relationship with local viewers," said Bob Marbut, chairman of Hearst-Argyle, whose stations reach about 18 percent of the United States.
Among Geocast's 75 employees are veterans from television and the Internet, including 12 from Netscape. The engineers integrating its software with existing Internet browsers, for example, include two who helped create the Web's first browser: Tom Paquin, employee No. 7 at Netscape, and Jon Mittelhauser, another Netscape refugee who co-wrote the original Mosaic software. Geocast's creative director, John Goecke, was the first art director for rival cable-modem service At Home Network, and its technical chief is NBC's former vice president for broadcast and network engineering, Charles Jablonski.
Geocast's programming vision is sketchy, but the idea is not to display full-length TV shows. After all, it is not a one-way, linear programming channel, but a two-way medium designed to be programmed by each viewer. For instance, it might combine video from traffic intersections on people's commuting routes with game clips from their favorite sports teams. Commercials would be full-motion video, shortened to 10 seconds and customized. "We believe that over time, this will create a whole new industry as we figure out how to program it," said Geocast chief executive H. Joseph Horowitz. "Initially, it's a little like learning how to use television after radio."
Geocast faces major challenges, not the least of which is people's willingness to buy yet another device. Yet it is an intriguing effort to expand the Net's architecture to make the medium more visually dynamic and compelling.
© 1999 The Washington Post Company