Cutting Through the Hype:
A few of the realities of Internet in Japan, and around the world
by Roger BoisvertWith this article I'd like to explain a bit of what the Internet is really like today, and what some of the major issues are, and a little on what the Internet will be like tomorrow. Internet providers in Japan have to spend a lot of time educating the consumer. My company has taken it upon ourselves to go the next step, and educate the consumers, not just on what the Internet could do for them, but also about what some of the realities are that no one wants to talk about.
It's been a long road from the days when I barely knew how to turn on a computer, and a lot of the thanks go to the TPC. The Internet services we created last year in September, in a way, are an extension of what I learned with the help of the TPC, and its members. During my 8 1/2 years with McKinsey & Company, amongst other things, it was my job to read the trends of technologies, and to use the understanding developed to guide our clients, especially those in computer technologies, toward future strategies. Also part of the job was to assist in the creation of our own technology use and future, both in Japan, and globally. That was a great exposure to the needs of companies, and of individuals. Having to work on a global basis introduced me quickly to the needs for communications.
In the early days, I was pretty much on my own in Japan as far as learning about, and using new technologies. It was then that the TPC first became a part of my life. Here, in Tokyo was a group of people, some as unskilled as I, and most willing to help answer questions, even beginners questions. One of the early, burning needs was to be able to communicate with people, not only in Tokyo, but also in other parts of the world.
It is here I first learned a little of the abilities of a modem and PC-based communications. There were people who needed to be talked with, and they were not in the office. Had to wait till 9 a.m. THEIR time to be able to talk with them, and of course that meant staying up late, or (shudder) having to get up really early. By using some of the same technologies as the TPC BBS has had now for many years I was able to communicate with these people in a way that was a lot more reasonable to me. Using the same skills learned on the BBS we found ways to move data and information more easily then before.
Over the years I did more and more work with the firm, and with the TPC in communications areas. Having served in most other areas of the club, at one point in the club's history there came a need for a new Sysop. That job, and I gravitated together, and lo!, a new Sysop was born. From there, perhaps it isn't so hard for people to understand some of the driving forces that led me to establishing the first licensed commercial Internet provider in Japan, and then, placing an even bigger bet, funding my own company, and creating another new Internet provider in Japan.
The following is an account of some of the "insider" events leading to the creation of the first commercial Internet in Japan. And then a little more about the trials of creating the new one. This follows with some thoughts to the changing future of the Internet. This will be the first of several installments covering these topics, spread out over several issues of the TPC newsletter (yes Mike, wipe that smile off your face).
The world changes quickly. The Internet changes even more quickly. People involved in providing Internet services say that the Internet two years from now will be totally different from what it is today. This is NOT an overstatement. The "academic" Internet in Japan was created by Jun Murai a few years ago. The commercial Internet in Japan began only in Sept., 1993...16 months ago. People are fond of saying that Japan is five years behind the US in computer networking. This may or may not be true in other areas of computer use in Japan. This article won't argue that one way or other. I've also heard it said that Japan is five years or more behind the US in Internet use. Hogwash! There are problems, and issues galore, but we are not that far behind now.
The commercial Internet was born in the US only three years ago, two years before we brought it to Japan. Since that meager introduction we have seen IIJ obtain its license, and after that thousands of companies and people gaining access to the Internet for the first time. Where there were three network providers one year ago, IIKK, IIJ, and SPIN, there are now at least a dozen providers, with several more scheduled to begin in January 1995.
The market in Japan is developing very differently than in the US. The Internet was born of a need for the US military to develop a network that allowed American Scientists to work on important projects, sharing from around the US some very expensive supercomputers and systems. Another objective of the system design was the ability of the system to literally withstand a nuclear attack. If the system was centrally controlled with a single network operating center, and that center came under attack, then the whole network could be destroyed with a single act of violence.
As a result, a distributed system was designed which could automatically reroute in case of difficulties. Those same features still exist in the Internet today. If one piece of the network were to go under, under most circumstances, the system would automatically route around the problem.
At the level of a large provider such as Sprint, this is quite true. For small providers, with only a single connection point, and lines going only to there, a failure at that one point for them means lost connectivity. In places where the costs are low it is less likely that anyone except the newest and smallest network providers would have that issue. In Japan, where costs can mount to 60 to 100 times as high as in the US to provide the same service it is a very rare provider who has more than one Internet access line linking their system with the Internet in Japan.
Come to think of it, NONE of the commercial Internet providers do. SPIN has a single 256k line. IIJ has a single 768k line. Those are the only two providers in Japan who carry their lines internationally. EVERYONE ELSE takes their feeds off of these two suppliers, except for the few people who use international dial-up lines to pick up a minor trickle of information for their BBS-based systems. This also does not include large companies who lease lines internationally for their own needs, but, don't provide access to other companies.
Let's put it this way, as of January 1, 1995 there are only two Internet providers carrying all commercial Internet access for all of Japan. The total bandwidth of commercial Internet in Japan as of Jan. 1, 1995 is 768k + 256k. This is less than a single T1 line, 1.5meg/second. In the US a beginning, underfunded startup commercial Internet provider commonly starts with a T1, perhaps going to their bedroom. I know of a person in the US with a T1 for private use. The cost of that T1 line is about US$1,000/mth. The cost of the same T1 linking to the commercial Internet, and carried all the way to the US would be almost ¥7 mil/month; i.e., 70 times as expensive as in the US.
Do you wonder why Internet access costs in Japan are priced somewhat above American costs? Wonder no further.
QUICK! NEWS FLASH!!! On January 10th, 1995 Global OnLine is having a small, 128k commercial Internet leased line installed, and that line is being leased all the way into the US to avoid the Japanese bandwidth bottleneck entirely!!
(Why? So we can offer our users high-speed throughput with modems ranging from 2400bps to v.34 28.8k on analog lines We are also providing full services to a limited number of leased line customers, and dedicated dial-up customers. The entire system is designed to deliver throughput customers pay for.)
Next question. Why lease a T1 all the way to the US when you could lease the same line to a local provider? Answer: would you pay for a 1.5meg line and be satisfied with only a 256k access line going international? No wonder bandwidth is limited in Japan, it costs an arm, and a leg, and ..... Ouch, my wallet!
OK, OK, so who cares? People SHOULD care, but, most users haven't figured it out yet. If you cram 10,000 dial up customers over a 256k line, each person dialing in with an average 9600 bps modem, what do you think happens to the throughput? 2400 suddenly feels very fast. Does that happen in Japan? Sure does. In a way, how else can a provider cover the painfully high costs of leased lines in Japan for Internet access? There are other ways, but, the basic economics of the Internet business, and of the almost the entire telecommunications industry can be summed up in two words "overselling bandwidth".
Some users are willing to put up with slow access on overcrowded lines, and with systems that have too few modems. Partly they may simply not understand how little they are getting for their money. Other users are willing to pay a bit more for higher speed access, on v.34 modems onto less crowded lines. These, and questions of network architecture are important questions when trying to understand the differences between different access providers, especially when they are priced similarly.
Oh, did I mention network architecture? Heavy phrase. The meaning is simple in this case however. I'm not talking about the network a company may have inside a building, but how your Internet access provider links to the net, through whom, and who else also links to that provider, and how big the bandwidth is, and how much it is oversold. Through some basic research we find that among Internet access providers in Japan, the generous ones oversell their bandwidth between 15 to 25 times. Others provide even more crowded lines.
In the next issue we will explore the wonders of NTT pricing, and of government policy that affects the Internet.
© Algorithmica Japonica Copyright Notice: Copyright of material rests with the individual author. Articles may be reprinted by other user groups if the author and original publication are credited. Any other reproduction or use of material herein is prohibited without prior written permission from TPC. The mention of names of products without indication of Trademark or Registered Trademark status in no way implies that these products are not so protected by law.
The Newsletter of the Tokyo PC Users Group
Submissions : Editor
Tokyo PC Users Group, Post Office Box 103, Shibuya-Ku, Tokyo 150-8691, JAPAN