Projects and Prototypes
The future, folks, will slowly unfold
Anyone who’s picked up a newspaper in the past six months can’t help being aware that big moves are afoot to upgrade the communications infrastructure of the United States. Because some 40-plus years ago the cable industry chose to wrap a housing around TV spectrum and pipe video signals directly into people’s homes, it is now undeniably in an excellent position to provide a relatively inexpensive ramp from consumer homes onto the so-called “information highway” that will enable interactive television and a host of other consumer information services.
The Projects and Prototypes panel at Digital World brought to the same table two of the most discussed companies implementing interactive TV projects and/or prototypes — Viacom and Time Warner — with database leader Oracle and the First Cities alliance to discuss the virtues and drawbacks of such networks as they’re under consideration today.
HOROWITZ: ‘IT WILL TAKE LONGER THAN YOU THINK’
Ed Horowitz is a senior VP of Viacom International responsible for Viacom’s new business as it relates to technology. Known as “the techie at the top” of the New York-based company, Horowitz first conceived of Viacom’s first interactive TV project, a fiber-based network that’s being installed in the San Francisco Bay Area community of Castro Valley.
He started his presentation on a cautionary note à propos of the vast amount of wind being generated by companies promoting interactive television, and we assume this includes his own. People in his line of business as an engineer often get trapped into believing that “if only this next chip could be installed, the world would be a better place.”
Channels are expensive.>> He said he wanted to provide a “dose of reality,” and level-set some expectations as to what we can expect, and when. “This is a balancing act,” he said. “Today it costs $100 million to start one single channel.” When building a system with a capacity of 500 or more channels, he said, we must ask ourselves, “If we build it, will the consumers come?”
(We’d like to see some research done along these lines, comparing the cost of getting a digital interactive channel up on a headend-based system vs. that of the phone companies for getting a video/information server up and running. Maybe digital production tools have pushed these costs down as well.)
The next question we need to ask ourselves, said Horowitz, is how long this will actually take to deploy. Engineers, he said, build on a heritage of knowns. The technologies being used in these fiber systems today, such as compression and fiber-optic transmission, are actually 20-year-old (or more) research projects coming to fruition. Technology, he said, always arrives earlier than you thought it would, but takes longer to implement.
It’s about money.>> And why is that? Because, Horowitz said, “It’s not about the technology, but it is about the money.” The settop boxes being built for Castro Valley, he said, will cost $800–$1,000 each, a far cry from the incredibly hopeful (and probably disingenuous) “$300″ that some companies are claiming. The national infrastructure upgrade will require installing thousands of miles of fiber-optic cabling, costing from $12–$18 billion to rebuild the entire industry.
Obviously it’s not cheap to implement intelligent networks on a broad scale. How to finance the infrastructure upgrade has been a key component. That’s one reason, he said, that cable is re-architecting its system into neighborhood segments. Laying fiber to the neighborhood essentially allows a cable company like Viacom or Time Warner to deliver potential access to very high-bandwidth programming (such as video on demand and home shopping) without having to make what both he and Geoff Holmes of Time Warner consider to be an overkill investment in fiber to the settop before the market is ready.
The benefit for cable, he said, is that they’ve already got one high-bandwidth pipe into or near the home. Direct broadcast satellite, on the other hand, has virtually no infrastructure in place; the settop devices they’re developing cost $700 per device, which Horowitz totes up to be a $7 billion investment with an additional $1 billion just for the satellite. (He didn’t cite the source of his statistics on this.)
And the national cost estimate to replace copper telephone wiring with fiber is staggering. Telcos don’t have the “to the neighborhood” option in the same way that cable does. They could indeed install fiber farther down the chain toward the home, which would allow them to get high-bandwidth programming closer to the consumer. But the copper wire that must carry this programming the final mile to the home has nowhere near the capacity of coaxial cable.
Key words: Potential and “2000.” These are non-trivial numbers for everyone involved, said Horowitz, but for cable more so than the others, there is great potential for returns. What fiber provides is “more channels,” which means more distribution options for programming. Horowitz said he believes that only 17–20 million homes will be hooked up to an interactive network by the year 2000. Though cable’s present market for basic rate services is about three times that size today, Horowitz believes a 17-million-home market will be sufficient to force changes in the economics of TV to support interactive programming.
Viacom’s Castro Valley project, said Horowitz, is his company’s foot in the door toward helping this happen. The project will serve 17,000 homes with 2,000 nodes. Different types of services will be targeted toward different demographic groups of 500 households each. It will be an interactive, two-way system, Horowitz said, with “several thousand” settop units deployed by 1994 (though he didn’t say when in 1994). Viacom will provide on-channel security and smart cards that work in tandem with security to prevent taping.
As Time Warner has said many times as well, Horowitz stressed that Viacom’s edge is in no small part the powerful franchise it has with users via its brand-name products such as MTV and Nickelodeon.
TIME WARNER IS CHANGING THE BALANCE OF ITS BUSINESS
Geoff Holmes is senior VP of technology for Time Warner, and, like Horowitz, is the senior executive for new technology business development. He’s also chairman of the Time Warner Interactive Group, formerly Warner New Media, and he spoke primarily about TW’s Orlando project, which is building the full-service, two-way interactive electronic network that we’ve all been reading so much about.
It’s more than a showcase for new programming today, said Holmes. He believes that “as long as we provide services of today, only better, there’s lots of evidence” that people will buy. They already spend $16 billion a year renting and buying movies, and about the same amount on video games. So at the outset, he said, instead of making them channel-surf, Orlando will eliminate their VCRs and provide them with “a dedicated channel with personalized services.” (We’d like to know who had this “personal channel” idea first — it certainly caught on with all the speakers at Digital World.)
Supercharged system.>> To do so, Holmes said, TW will “supercharge” the system, spending 30 months laying fiber, installing ATM switches for two-way video, audio and data transmission, and working with Silicon Graphics and Scientific Atlanta to provide video servers that hold 1,000 hours of full-motion video and a settop box to deliver them.
The Orlando system is installing “fiber deep into the neighborhood,” as described by Viacom, above, to serve 300–400 households each. Holmes said TW’s Quantum architecture for channel compression will be deployed there, and in almost all its systems during the next five years.
No twitch games.>> He sees the Orlando project as the beginning of a network of networks that can connect to any other source of data, including online systems such as Prodigy, the electronic version of the L.L. Bean catalog, etc. The only programming type he thought might be a challenge for the network was “twitch games,” those Nintendo-type games where a kid beats on the controller as fast as he can to shoot (stab, maim, whatever) the opponent.
Though Holmes for some reason downplayed the telephony side of what TW is attempting with the Orlando project — it plans to provide local phone service via cellular and interactive TV programming — he did say a bit about how this kind of massive network deployment will change the balance of TW’s business.
Changing the balance.>> Right now, he said, 60 percent of TW’s revenues are from content. (It is the largest intellectual property/copyright holder in the world.) The other 40 percent is conduit, via its cable businesses, etc. “As this number improves,” he said, “we’ll become more involved in telephone kinds of businesses, such as 30-frame-per-second video telephony.” He also said that TW intends to grow its customer base out of the “homes passed” phase to include schools and businesses.
Time Warner is planning a software developers conference for those who want to provide programming on the Orlando system, but Holmes didn’t provide details.
THIS IS A LOT MORE WORK THAN WE THINK
The presentations of Bruce Sidran, executive director of the First Cities alliance, and Larry Ellison, chairman of the Oracle Corp., differed greatly from those of the two cable executives, but they certainly joined Ed Horowitz in spirit by saying, “Whoa! Hold on a minute” to some of the assumptions about how such networks will be deployed.
Sidran, at the helm of an alliance with a dozen members including Kaleida Labs, Comsat Video Enterprises, Apple Computer, Southwestern Bell and Tandem Computers, started his presentation by saying that “fear and greed are powerful motivators.” (For more on the First Cities alliance, see Vol. 2, No. 7, p. 3.)
Throwing technology.>> Although Sidran didn’t say it quite this way, it seems that First Cities doesn’t believe that selecting a test site and throwing technology at it is necessarily the most logical progression of events to pursue.
The idea behind First Cities was to get companies together and work out some of the problems that TW and Viacom are going to face on site when constructing a network to deliver multimedia services. The key goal for First Cities is not how or when to get the fiber in place, he said, but to come up with the interface specifications that will be necessary to create, deliver and display interactive programming on the network.
Accommodating hybrids.>> In addition, First Cities believes that such services will be delivered over a variety of different network types, not just via the cable system, and that work must be done to solve the technical — and social — problems of delivering information over hybrid networks as well.
Sidran just moved his office to Washington, DC, and is spending a fair amount of time discussing issues such as the proper role of the government in a global communications infrastructure and ensuring that critical issues such as multiculturalism aren’t overlooked in the rush to market.
YOU CAN’T JUST POINT A CAMERA AT SOFTWARE
Ellison echoed Sidran’s theme, at least on the technology side. Though Ellison was a bit disingenuous in saying that the work Oracle is doing with US West (to provide a database server for video) would be able to interoperate with every computer system in existence, he did make a very good point about the difficulty of building good, robust software.
Even the big guys say so.>> The kind of software that will be required in a full-service network — especially if Sidran gets his wish and Holmes’ “network of networks” includes telcos, satellite, PCS and cellular — is unbelievably complex, difficult to create and even more difficult to test and make foolproof in the uncontrolled and uncontrollable environment of the consumer living room. You can’t just point a camera at it and get a product out the other end, as Sun’s John Gage said. All of this digital wonderfulness will be much harder, and take much longer, than any of us can imagine.
Denise Caruso