NTT's Open Computer Network: Bad News or Good News for Japanese ISPs?

NTT's Open Computer Network (OCN) services, slated for launch early in 1997, will forever alter the Japanese communications business market. Proponents predict that, by enabling businesses and individuals to access the Internet at lower cost, it will h elp Japan make up for lost time in the competitive world of global networking. But some of Japan's larger Internet service providers view OCN as a threat.

by Noriko Takezaki

Open computer network services (OCN) is NTT's first attempt to provide a routing service for a LAN-to-LAN communications network by using TCP/IP (Internet) technology. A major goal of this "connectionless, multimedia service suitable for communication bet ween computers" is to promote easier and lower-cost access to the Internet in Japan.

Bringing down the cost of Net access

Use of the Internet has grown dramatically in Japan during the past couple of years. That growth could have been even more dramatic, however, if not for the high fees charged for leased lines in Japan -- about five times the cost for si milar service and bandwidth in the US. The cost to the provider is passed on to the consumer.

To improve this situation (one for which it is often blamed), NTT plans to develop OCN as a new network system that will transmit information via packets, node by node, until the packets reach their designated host. This "best-effort," connectionless netw ork structure, which differs from a traditional telephone network, will allow NTT to lower its service charges drastically. (The rates noted in the table on page 18 are tentative; NTT is expected to submit its OCN service charge plan to the Ministry of Po sts and Telecommunications (MPT) for approval in September.)

"We plan to offer service charges which can be lower than [those in] the US for low-speed services, and in a similar price range with the US for the higher speed services," says Kazuaki Katori, NTT's vice president and executive manager in charge of the O CN effort. "OCN is quite different from our conventional end-to-end connection type network, which provides high-performance communications services by guaranteeing a given bandwidth, but requires a higher service charge because of such performance."

Proposed OCN services

Within the OCN scheme, three basic services are planned. OCN's Enhanced Services for Integrated Corporate Communications, as the name implies, is intended to meet the needs of corporate customers. These services include multicasting, security, virtual LAN , directory/navigation, and resource reservation services that are appropriate for intranets and electronic commerce applications. And the proposed Functions for Information Providers and Individual Users will support the use of network applications, part icularly for electronic commerce, by information providers and individual users. Included will be such services as authentication, security, financial settlement, and server rental.

Among the three planned services, though, it is the OCN Circuit Service that will have a major market impact. For a relatively inexpensive flat rate, all communications carriers and Internet Service Providers (ISPs) that hold a Type I or II license from t he MPT will be able to use both the OCN backbone and OCN access lines. The OCN Circuit Service is available either through dial-up access via the Public Switched Telephone Network (PSTN) or through Narrowband ISDN (N-ISDN) at speeds of 128K, 1.5M, 6M, and 50M bps.

The OCN Circuit Service for backbone poses a significant threat to some ISPs, particularly the larger ones with a Special Type II license who offer corporate services based on leased lines (such as Tokyo Internet, IIJ, and C&C Mesh Net). Tokyo Interne t, for example, currently sells its 128K bps service at ¥250,000 per month; NTT could offer equivalent OCN-based service to the same customer for ¥30,000 to ¥40,000 per month, less than one-sixth of Tokyo Internet's price.

"To compete with the OCN [backbone service] on price, the only thing we can do is increase our number of subscribers in order to lower our service charge," said Toru Takahashi, president of Tokyo Internet, at a seminar held in Makuhari Messe in July. But this would not be easy, he added, since Tokyo Internet would have to increase its number of subscribers fivefold to be able to match the projected OCN pricing. Takahashi noted that an industry organization of Type II carriers has already lodged a complain t with the Telecommunications Committee, an MPT advisory body. OCN, they charge, will allow NTT to monopolize the local network market, thus hindering the fair competition that the service is supposedly intended to promote.

Some ISPs are doing more than complain; they're fighting back. Internet Initiative Japan (IIJ), for example, has announced a tie-up with cable TV operators that seems designed to compete with the OCN backbone service. In cooperation with about ten cable T V operators in Japan (including Tokyo-based Jupiter Telecom, Mie-based CTY, and Aichi-based Himawari Network), IIJ plans to start an Internet access service that uses the cable TV networks in October. Since this new service bypasses NTT's network, IIJ is expected to be able to set service charges that are competitive with OCN's planned charges.

Opportunities and concerns

Second-tier ISPs, meanwhile, are generally positive about the OCN Circuit Service, since their cost for using the service seems likely to be much lower than that of their current infrastructure (often foreign carriers' leased lines). "OCN will... ease our financial burden for the infrastructure," says Craig Oda, president of TWICS. "If the quality is good, we will probably switch [to OCN]. We are always looking for high-quality infrastructure providers that can offer good service and have good policies as to how many customers they put on a trunk."

Indeed, the major concern of many about using OCN is, as Oda mentions, the communications quality it can provide. As a best-effort type, low-priced network, OCN offers no guarantee about transmission quality, and thus will not be suitable for many critica l business applications. In addition, industry researchers warn that, in terms of service quality, TCP/IP technology cannot ensure error-free transmission. The loss of IP packets is common when the network traffic is heavy, which can lead to deterioration of communications quality (particularly for image data).

TCP/IP may, in fact, be only a temporary solution for fast introduction of the new network; many expect the protocol to be replaced by asynchronous transfer mode (ATM) technology in the near future. Asked about this point, Masaya Okada, NTT's manager in c harge of OCN, admits that they are considering the possibility of using an ATM layer for their 50M-bps service. He cautions, however, that many aspects of ATM technology are still largely unproven, so time will be needed to evaluate it thoroughly and deci de whether to apply it to OCN.

Aside from the technological aspects, there is genuine concern about just how "open" NTT will be in its policy toward offering access to Type II carriers. NTT says that, for OCN, it will allow Type II carriers to place their own routers and transmission e quipment inside the NTT operating facilities, and that it will offer a special leased line service charge for connecting the routers and equipment inside NTT. Unfortunately, NTT has a bad reputation in this regard based on its past interconnection "negoti ations" with Japan's new common carriers (NCCs); NTT charged the NCCs high interconnection rates and did not allow the interconnection in a timely manner.

Many Type II carriers, therefore, are concerned that they could face a similar interconnection dispute regarding the "wholesale" price of OCN access lines and the placement of their equipment in NTT facilities. "We hope NTT will propose a reasonable servi ce charge, and reasonable conditions [for the placement of equipment]," says Tokyo Internet's Takahashi. The unvoiced "but we'll believe it when we see it" on the end of his statement was obvious.

Taking the initiative

Aside from OCN, NTT has been promoting "opticalization," or Fiber-to-the-Home (FTTH), to realize a nationwide infrastructure. NTT has almost finished converting its trunk lines to optical fiber, but the conversion of access lines has been delayed by doubt s as to whether the benefits to subscribers will justify the expense (and concomitant increase in service fees).

It is thought that NTT thus developed its OCN plan, in addition to the continuance of the FTTH plan, as a quicker solution for creating an open nationwide network -- one that can keep pace with the rapid growth of computer communications. The service char ge for LAN-to-LAN communications via OCN will be only about 2% that required for leased optical fiber access lines.

For NTT, OCN represents a big step forward in the transition from conservative telecommunications carrier to flexible multimedia service provider. "We often say, 'Don't stick to the viewpoints of a telecommunications carrier any more,'" observes NTT's Oka da. "We believe that, through the introduction of the OCN, we can make a great contribution to the era of multimedia communications in Japan."

Whether NTT's vision of OCN can be realized will depend on how objectively the megalithic carrier can see itself and its competitors, and understand their roles. The Open Computer Network represents a turning point not only for the telecommunications mark et in Japan, but could also give a competitive boost to Japanese corporations in the midst of today's global communications revolution.