Who will wire the nation?

Three years ago, most people expected that the job of moving digital data around the country would naturally fall to the phone company. But the slow rollout of the Integrated Services Digital Network, or ISDN, as well as its half-hearted backing of upgrading American homes to high-capacity, high-speed fiber-optic phone cables, has led to doubts whether the telcos have what it takes. At the “Cable, Cellular and Satellite: The ‘Other’ Information Networks” panel it was clear that other institutions — cable operators and satellite companies — have begun to stake their claims to the new digital territory.

Cable ups the channel count. Terry Hershey, a business planner from the executive vice president’s office at Time Warner in New York, described the new Quantum cable system going into the Borough of Queens, NY. It will replace the old coaxial distribution network, with its branching tree topology and many amplifiers. The new system uses a “fiber to the neighborhood” approach: each neighborhood (comprising 500–1,500 homes) is served by a dedicated optical fiber that runs directly from the cable headend with no amplifiers.

Within the neighborhood, the signal is still distributed to each home by coax; but no home is more than four amplifiers from the end of the fiber. Quantum should thus be vastly more reliable than its predecessor. It also becomes feasible to offer two-way communication between each subscriber and the headend office.

The fiber will have a 1-GHz bandwidth. Half of that bandwidth will be used for 75 ordinary analog (but digitally encoded) television channels, essentially replicating the service that the old system provided. The other half will be used for pure digital service. Using digital compression to squeeze three channels into the space usually occupied by one would be sufficient to add 200-plus channels, but Hershey expects that by the time Quantum is complete, it will be possible to compress TV programs by as much as 8:1, allowing more than 500 channels.

(See “Interactive TV Arrives” story, p. 20, for information on Viacom’s upcoming fiber-wired cable system in Castro Valley, CA.)

Most of these channels will be invisible to the user, being used for such services as pay-per-view, which would be selected from a menu of movies and start-times. A Broadband Option Selector System, or BOSS, would route the appropriate channel from the fiber to each subscriber.

With such a high ratio of channels to subscribers, it becomes possible for each subscriber to have his own channel for interactive services. Time Warner is exploring potential business opportunities for delivering such services, including:

• Subscriber-to-subscriber telecommunications (Time Warner has an experimental license from the FCC);
• Entertainment and games;
• Shopping (coupons to the home, preference profiles on each viewer);
• Financial and professional services;
• Information services; and
• Household services.

Users would get the benefit of a wide range of selections, while advertisers could target their messages to single homes.

Direct broadcast satellite. In some parts of the world, direct broadcast satellite transmission already has greater market penetration than cable. Sky Pix (among others) would like to see the same thing happen in the U.S. with direct transmission of compressed digital video to subscribers’ homes.

Bob Kniskern, chief scientist for SkyPix, described how SkyPix’s satellite distribution scheme works. The advantage of satellites is that the signal is available everywhere, today, even out in the boondocks where cable will never go. There used to be two disadvantages: the cost and size of the receiving equipment, and the fact that transmission goes only one way. But things have changed.

Right now, SkyPix intends to sell the hardware for a complete receiver setup (dish and digital decoder) for about $850. (Besieged by a number of business problems discussed in the June issue of Digital Media, SkyPix is now about three months late launching the service.) Installation is sold separately.

The receiver setup includes a 36-inch dish — small enough not to run afoul of most zoning ordinances –and a TV-top decoder box. The decoder, though it looks simple to the user, is a fairly smart computer that can be completely reprogrammed by downloading software from the satellite. It has a 250-channel capacity; each channel uses 3.375 megabits/sec. of bandwidth, which supports compressed video and audio, control information and error-correcting codes.

As for interactivity, SkyPix already maintains an 800 number for the return link from the decoder back to its billing computer, so the potential is there. The phone line has low bandwidth, but it is more than sufficient for selecting shows or for electronic polling of the sort presidential hopeful Ross Perot has been touting. SkyPix markets the service as a kind of “rental store in the sky,” but is looking at offering electronic publications and education for remote sites as well. To truly compete with cable, it may have to offer popular cable programming (CNN, MTV etc.) as well. Kniskern had no comment on how — or when — SkyPix would be able to do this.

Cellular news transport. Both SkyPix and Quantum are oriented toward home entertainment, but Steve Webert, vice president of product development at Oracle Data Publishing in Seattle, WA, described a business-information service that Oracle and McCaw Cellular are jointly developing.

The plan is to gather data from government sources, news wires and so on; to transport it worldwide by satellites, land lines or whatever carriers are handy; and if necessary to rebroadcast it to so-called “nomadic” devices. Business information, being largely text and numeric data, needs much less bandwidth than video. Webert thinks that a standard T1 channel would be more than sufficient for the uses he envisions, and lower-capacity technologies will have to be used during the startup phases.

‘Cents per month.’ There will be just one data stream, from which users can extract (and pay for) whatever they are interested in. Customers could even temporarily store information for later acceptance or erasure, paying only for the data they keep. Oracle’s goal is to become “the Henry Ford of data,” according to Webert, so it will price the information as low as possible, with cut rates for regular and high-volume customers.

The cost, Webert said, would be “measured in cents per month.” Oracle would make money on volume; the market today, he thinks, is but 2 or 3 percent of what it could grow to over the next 20 years.

Are the telcos doomed? Does all of this spell doom for the phone company? Robert Rosenberg, president of Insight Research Corp. in Livingston, NJ, thinks that it would be a mistake to underestimate the telcos. For one thing, there is new switching technology coming over the horizon; in 20 years, monolithic optical switches could offer both point-to-point dial-up access and enormous bandwidth.

For another, the current rate of replacing copper wires with optical fibers is the direct result of depreciation regulations. With different rules, the phone companies could install fibers to homes on a pay-as-you-go basis. Finally, the cable operators and the phone companies might join forces, allowing the phone companies to contribute their skill in administration and billing.

Mike Grubbs, senior director of marketing for Tandy Corp. in Fort Worth, TX, points out that it is far too soon to settle on one digital distribution system. Each of the existing pipelines is optimized for a certain purpose and will probably continue to exist for a long time. Tandy, at least, intends to support many different data pipelines. Indeed, this seems like the only intelligent path to pursue.

Peter Dyson