Cable Meets the Internet

A chance for Japan to lead the world, or just a big waste of time?

If you follow industry news from the States, you're familiar with the hubbub over superfast cable modems that claim 10M bps (10 megabits per second) of bandwidth to your home computer and promise to revolutionize Web surfing once and for all. Unfortunately, the current "one-way" cable TV infrastructure - 10M bps to the home, but only 9600 bps out - is ill-suited to the task. Plus, it is not a very clean signal; television can hide a lot of noise (and, as we all know, keeping a link to an Internet provider isn't always easy).

Not to be confused with the concept of Internet TV (like WebTV), which uses regular phone lines, today's cable Internet proposals rely on cable TV coaxial wire. A special cable modem connects to a local cable switching station, which in turn connects to an Internet provider or a cable-Internet network.

The general definition of "cable" simply specifies that signals are received through wires as opposed to via large antennas or satellites. In the near future, we will start to see cable Internet service refer to other types of wires as well.

I've often thought that, because the nation lacks a mature coaxial cable infrastructure, Japan is in a good position to create one that allows fast two-way data connections (thus leaping ahead of the West, which must retrofit existing networks). Or, if there is enough momentum in alternative technologies, Japan could even skip the coaxial-wire generation and invest in the future, such as fiber-optic cables. As usual, though, it seems that Japan is hedging its bets and experimenting with a variety of systems.

Infrastructure projects
There are several large ongoing pilot projects; almost every CATV operator in Japan is conducting tests or feasibility studies of delivering Internet packets over their networks. These projects often are not widely publicized, but some of their overseas purchases can be a tip-off.

Japanese cable operators are teaming up with a variety of US manufacturers, especially when it comes to cable modems. In December 1996, for example, Newsbytes reported that a group of four Japanese cable companies will jointly purchase cable modems from Hewlett Packard. The operators - Tokyu Cable Television Co., Kintetsu Cable Network, Himanari Network Co., and LCV Co. - represent about 300,000 subscribers. Newsbytes also reported in December that California-based Hybrid Networks is teaming up with Japan's Sharp Corporation and Itochu Corporation for the design, manufacture, and distribution of cable modems suitable for both telephony and data transmission.

In October 1996, Newsbytes reported that Marubeni Corporation has chosen Motorola cable modems for use on its network in Kanazawa. Although the network had only 33,000 subscribers at that time, it is accessible to over 300,000 households, making it Japan's largest broadband network.

NTT and Microsoft, meanwhile, are in the final months of a yearlong video-on-demand system test. As many as 300 houses have been wired to the network by fiber-optic cable and receiving transmissions as high as 155M bps.

A hint of the future
NTT says that, beginning in July 1997, it will conduct a series of trial services based on fiber-optic cable, mixing cable television channels and services with public telephone service and other NTT technologies. And sometime this year, Japan's Ministry of Construction plans to start laying roughly 150,000 kilometers of fiber-optic cable along the nation's roads and expressways. When completed, in 2010, this would be a completely independent network available to all users (and would possibly compete with NTT's fiber-optic network). Cable operators, local telecom companies, and Internet backbone providers are expected to be the primary customers.

Japan received a much-needed boost in overall bandwidth to the outside world when the new TPC-5CN undersea cable was completed earlier this year. This transPacific fiber-optic cable makes a giant round-trip loop between Japan and the US and adds over 10 gigabits of bandwidth to voice, data, and video communications. The cable connects into all the world's major networks.

All of the bandwidth in the world, though, is not a great advancement without compelling content. While viewing Web homepages at 10M bps sounds very appealing, it also makes it easier to get bored faster. Interesting content is the key to the implementation of a fast infrastructure.

Lots of things are being considered, like video-on-demand, network gaming, video and audio streaming, and project collaboration. Pilot tests so far, however, have returned disappointing results on the much ballyhooed video-on-demand. My personal prediction is that network gaming will take off in the home, and project collaboration will be the killer application for corporations. Both of these will benefit greatly from a new generation of high-speed Internet connectivity.

Send your comments or questions directly to Forest at, or via the editorial staff to (fax 03-3499-2199).