Article on Wiring Japan from Wired several issues ago. Worth subscribing to!! The next Japan will co

[David Farber <>]

-----
Wiring Japan


In the first of an exclusive two-part series, WIRED's Japan correspondent
reports from Tokyo on the bitter culture clash that has reduced Japan to a
third-rate power in networking.


By Bob Johnstone


At precisely 8:00 on the evening of Friday, September 17, 1993, Japan's
first commercial Internet packets flashed out of Tokyo and down
Trans-Pacific Cable No. 4, bound for San Jose, California. A new era in
Japanese networking had begun. As befits the birth of a new business,
cheers went up and toasts were made. But not everybody was rejoicing in
Tokyo that night - for Japan's first commercial Internet packets were sent
by American engineers working for Japanese subsidiaries of the US
corporations InterCon Systems and AT&T. InterCon's first customer was
TWICS, Japan's first public access Internet provider, a small for-profit
firm most of whose 400-odd subscribers are foreigners based in Japan.


Across town, a group of Japanese Internet pioneers were grinding their
teeth in frustration. The company they had set up to provide commercial
Internet services had been denied a license to operate by Japan's Ministry
of Posts and Telecommunications. Holding up the locals while waving on the
foreigners is not the way business is usually done in Japan. Something odd
is going on.


That something, in essence, is the head-on collision of two cultures: The
freewheeling, democratic style of the Internet has run smack into
traditional Japan at its most authoritarian. On one side, you have the
technology pioneers, young volunteers who built Japan's largest research
network by their own efforts, without any support from the Japanese
government. They are led by Jun Murai, the man some Americans (like Carl
Malamud and Howard Rheingold) call the Internet samurai.


On the other, you have the officials charged with providing network
services to the Japanese research community. They have tried to ram
unpopular standards and technolo-gy down users' throats - and failed. Their
leader is Hiroshi Inose, arguably Japan's most powerful technocrat.


The officials resent the pioneers' early successes and are waging a
dirty-tricks campaign to try to regain the upper hand. Through their
arrogant behavior, the pioneers have played into hands of their rivals, who
are masters of the bureaucratic game.


Today, the situation has degenerated into a highly emotional conflict, with
each side hurling accusations and insults at the other. Little TWICS has
been caught in the crossfire. In mid November, the company's long-standing
domestic e-mail connection via Tokyo University was suddenly cut off,
apparently in retaliation for TWICS having opted to use a non-Japanese
Internet link. Then the company received an intimidating phone call from a
man claiming to represent the computer center at Tokyo University.


"Stop doing business in Japan!" the man shouted, "Shut down at once!"
(TWICS has since been reconnected.) "It's one of the trickiest messes I've
seen in years," comments Internet luminary David Farber, a University of
Pennsylvania professor who tracks developments in Japan. It is also a mess
that matters. For, as Farber points out, what the Japanese do affects the
rest of us. And while Japan may be the world's second-largest economic
power, the Japanese remain dangerously isolated. Networking has the power
to change that by bringing Japan closer to the international community. But
by the same token, failure to log on to the world's largest network could
leave Japan more isolated than ever.


The massive proliferation of the Internet has left the Japanese far behind.
As of June 1993, Japan had roughly five networks for every 100 in the
United States. Outbound NSFNet traffic from Japan that month was 42,000
Mbytes, roughly the same as that from Taiwan, a country with one sixth
Japan's population, and less than half that from Australia, the Pacific
Rim's most aggressive network user.


Young Japanese have heard about the Internet and they are eager to get
access to it. The irony is that the very people who should be encouraging
them to log on are instead preventing them.


The best way to reach Jun Murai, associate professor at Keio University,
is, surprisingly, not by e-mail. Instead, you ask one of his acolytes to
track him down for you. Initial contact with Murai - via his car phone - is
encouraging: "You want to do [the interview] over a beer, or dinner, or
what?" he asks. On meeting Murai, you quickly realize why he is so popular
with his Internet counterparts elsewhere. In a country where most academics
still wear suits, Murai wears an ancient sports shirt, a beer gut tumbling
over his black jeans. He looks a bit like a bear, an impression his deep,
rumbling voice reinforces. And while vagueness is regarded as a virtue in
Japan, Murai comes straight to the point.


Ten years ago, when Murai was just 28 years old, the Japanese research
community was debating how to take advantage of deregulation to install
communications networks. Discussion centered around which of the proposed
Open Systems Interconnection architectures to adopt. To Murai, such
meetings were a waste of time: "I was young, and that was boring," he says.
"What I wanted was to have a network, to do actual operation and
development, so that we could find out what the problems related to
computers and communications were, then solve them."


So he and some friends rolled up their sleeves and started laying cable.
Their first effort was the immodestly titled JUNET, a dial-up modem service
offered over public phone lines. It proved immediately popular with
Japanese academics starved for e-mail, especially after Murai made it
possible for them to enter text using Japanese characters. Encouraged,
Murai went on to launch a more ambitious project in 1987, the Widely
Interconnected Distributed Environment, or WIDE. A backbone network, WIDE
is based on leased lines that interconnect local area networks. Owning
leased lines is very expensive in Japan, and to pay for them, Murai turned
to commercial firms like Sony and Canon. Once again, the network proved
popular - today, it connects some 30 research institutes and 40 companies.
WIDE also has a link, via the University of Hawaii, to NASA's Ames Research
Center, through which Japanese researchers can communicate with their US
counterparts.


Indeed, WIDE became such a hit that Murai was forced to turn down requests
for connections. A second headache was that companies were not necessarily
confining their use of the network to research, as required under the
Japanese government's extremely strict definition of appropriate use. An
obvious solution to both problems was to set up a company to offer
commercial Internet services. In December 1992, Murai and some of his
students formed Internet Initiative Japan (IIJ). This is the company to
which the Ministry of Posts and Telecommunications refuses to grant an
operating license.


Jun Murai's success is a thorn in the flesh of Shoichiro Asano, a former
Tokyo University professor who is in charge of day-to-day operations at
Japan's National Center for Science Information Systems (NACSIS). This
organization was formed to provide network information services for
university researchers. As an official organization, supported by Japan's
Ministry of Education, Science, and Culture, the Center naturally opted to
use officially sanctioned technology, notably the Open Systems
Interconnection protocols, which were first proposed in the early 1980s.
Trouble was, it took a lot longer than originally anticipated for the
committees in charge of developing these protocols to come up with the
goods, and even when they did, many found them cumbersome and needlessly
complex.


Meanwhile, back in the US, an ad hoc set of protocols known as TCP/IP was
spreading like wildfire. Establishment engineers turned up their noses at
these protocols, sniffing that they had been designed by young cowboys and
were too sloppy for any self-respecting network to use. Maybe they were -
and the protocol issue still divides the engineering com-munity with all
the ferocity ofa religious schism - but by the beginning of this decade,
TCP/IP had become the de facto standard for networking in most of the
world. It now has an installed base three to five orders of magnitude
larger than that of Open Systems Interconnection. In 1991, realizing that
it had backed the wrong horse, NACSIS at last began to convert its network
to support TCP/IP. By that time, however, WIDE was firmly established in
the eyes of the Internet community as Japan's front-runner.


That a lowly assistant professor and his rag-tag band of graduate students
should have attained such status seems to have embittered Asano, a
disheveled and shifty-looking individual in his early 50s. Mention Murai to
him and he becomes animated. "For five years, Jun Murai has done dirty
things to [the Center]," he complains. Many of these "dirty things" concern
Internet Initiative Japan. For example, Asano accuses Murai of being its
"shadow leader." (Japanese academics are not supposed to sully their hands
through contacts with business.) He also charges that IIJ has attempted to
use American pressure in various forms to force the Ministry of Posts and
Telecommunications to issue the group an operating license. None of this
would matter much if it were just a case of sour grapes. Unfortunately,
however, Asano has the backing of the director of NACSIS, Hiroshi Inose;
that gives him the power to do Murai - and, by extension, the spread of
Japanese networking - considerable harm.


In a society that reveres seniority, the 67-year-old Inose is about as
senior as you can get. Back around 1957, as a young electrical engineer
consulting at Bell Laboratories, Inose won a basic patent on time division
multiplexing, a key technology for combining several calls on the same
line, one which all modern telecommunications switching systems use. This
reflects a caliber of achievement that few Japanese academics can match. In
later life, Inose became dean of Tokyo University's prestigious engineering
school, taking up his current position as director of NACSIS upon his
retirement from the university. He is a member of the US National Academy
of Sciences, an IEEE fellow, and has a string of other awards and prizes to
his name. Prime among them is his designation by the Japanese government as
a person of cultural merit, an honor carrying tremendous status in Japan.
Today, Inose chairs many of the key policy committees at both the
ministries of Trade and Industry, and Posts and Telecommunications. His
former students hold senior positions at leading Japanese electronics
firms. Indeed, it is scarcely an exaggeration to say that few Japanese
involved in information technology are not beholden to Inose in some way.


Inose was not available to be interviewed for this article: His schedule
was booked up for two months. But most people who have met him are struck
by his charm and gentility. Hell, the man even writes poetry.


But inside the velvet glove is an iron fist. Many Japanese criticize Inose
for throwing his weight around, but few dare to do so openly. One of the
few is technology journalist Yukihiro Furuse. In the November issue of the
monthly magazine Shincho 45, Furuse writes that "probably no one can
criticize [Inose] because he has absolute power and he is senior to
everybody." Furuse recognizes that networking is of crucial importance to
Japan's future. And he is concerned that Inose's behind-the-scenes style is
not the best way to promote its growth.


One specific charge made by Furuse and others against Inose is that he has
used his influence with the Ministry of Posts and Telecommunications to
prevent IIJ from getting an operating license. Another is that he leaned on
the Ministry of Trade and Industry to prohibit a new computer research
project it was supporting from using IIJ to provide network services. Why
should Inose care about a bunch of young upstarts like Murai et al?
"Professor Inose represents the old establishment," suggests one senior
telecommunications executive, "and their idea is that government and
national universities should always stay in the center, delivering and
exchanging information." In the US, the acceptable-use policy for the
Internet was designed to promote the growth of commercial Internet
services. The Japanese acceptable-use policy seems by contrast aimed at
preventing the growth of Internetworking beyond the ivory towers.


David Farber, who has known Inose for many years, insists that his friend
is acting purely in the national interest and for the benefit of the
education community. Farber says that Inose is worried about IIJ's claims
that it can provide network services for Japan's university researchers at
an affordable price. In addition to rendering NACSIS superfluous, this
could also lead to the education ministry cutting off much-needed financial
support for networks. An IIJ spokesperson confirms that the company had
planned to offer an academic discount plan but were told by the Ministry of
Posts and Telecommunications that licensed carriers would not be allowed to
discriminate between different types of customers. IIJ is currently
providing customers with domestic network service, which it can do without
a special license.


"Inose believes that achieving a successful conclusion is a slow, careful
process that keeps the support of the education ministry," Farber says. "He
gets upset with the cowboy approach" taken by Jun Murai and his associates
at IIJ. Inose is not the only one upset by IIJ's behavior. Some of the
blame for the current conflict must also go to IIJ's manage-ment, and in
particular, to the company's president and CEO, Hiroyuki Fukase. Fukase is
an engineer by training, not a businessman. He has needlessly antagonized
potential investors by asking for money, then neglecting to perform the
follow-up visits that Japanese business etiquette demands. Worse, he has
annoyed Hikaru Chono, director of the Ministry of Posts and
Telecommunications' computer communications division and the official in
charge of issuing the license IIJ needs to provide international Internet
services.


In Japan, it pays to treat bureaucrats with respect. They form, after all,
an elite, and have dedicated themselves to serving their country, working
long hours in overcrowded conditions for low pay. The quid pro quo is
power. Visit a Japanese government office and you will see a constant
stream of supplicants who come to beg for official favor.


Chono is a prime example of the bureaucratic breed. A graduate of Tokyo
University's law school, he has paid his dues - including a stint as
Japan's representative at the CCITT (the international telephony committee)
in Geneva, and a year as a postmaster in a remote part of northern Japan.
He has been involved with telecommunications since 1975, and he is tired of
it. In particular, he is fed up with Fukase. "IIJ is a very difficult
company for me to understand," he sighs, "they've done lots of publicity
and marketing, they've published their tariffs in journals and magazines,
but they haven't finished the formalities."


What Chono needs is a letter of guarantee from a financial backer. "If they
complete all the necessary documents," he says, the Ministry of Posts and
Telecommunications will issue a license "within 15 days, maximum." Fukase
claims to have found a backer, the Industrial Bank of Japan, but accuses
Chono of having warned the bank not to deliver the letter of guarantee.
Such accusations are typical of Fukase's bull-in-a-china-shop style. In
early 1993, he upset the ministry by arranging for a letter from a friend
in the US National Science Foundation in support of IIJ's license
application to be delivered via the US Embassy in Tokyo. Asano accuses IIJ
of urging US equipment manufacturers to criticize the Ministry of Posts and
Telecommunications for preventing them from doing business in Japan. Such
heavy-handed attempts to use US pressure to bully the ministry into
granting a license have had an effect oppo-site to the one intended. The
best thing that Fukase could do at this point would be to go see Chono and
try to pour some oil on troubled waters.


But Fukase has not been to Chono's office in months, while Asano reportedly
visits him regularly. "They're naive guys," comments one observer of IIJ.
"They don't know how to play the game in their own country." Murai would
like to wash his hands of the conflict and get on with his research. "I
don't want to deal with any of this," he groans, adding in frustration,
"What's wrong with me? I'm providing a better environment [than NACSIS],
producing researchers and good results, and the com-panies [that support
WIDE] are very happy. The problem is," he concludes, "I'm too young to deal
with this kind of thing."


But much as he might like to, Murai cannot simply walk away from the mess
that surrounds IIJ. Indeed, as Farber points out, much of the current
problem stems from confusion over Murai's dual role as director of an
academic research network and would-be godfather of a commercial Internet
pro-vider. Carl Malamud, a friend of Murai and author of the technical
travelog, Exploring the Internet, comments that "the real issue in Japan is
the same as in the US. We're moving beyond the myth that the Internet is
some academic research project."


What then is to be done to sort things out? Nobody in Japan seems to know.
One less-than-optimum solution is simply to let time take its course. Inose
is expected to retire in a couple of years. Following his master's
departure, Asano says he will probably return to academic life. Well before
then, IIJ will likely bring in new management to run its business, leaving
Murai to get on with his research. For the moment, however, though a new
era has begun, precious time is being lost. And it is time that Japan can
ill afford to lose.




SIDEBAR
Reasons for Japan's Late Start
* Dominance of centralized, mainframe-based computing
* Lack of LANs
* Proprietary protocols
* TCP/IP ignored while government and industry pursued OSI
* Lack of Japanese software and support for routers
* Over-regulation
* Difficulty of Japanese text entry
* Japan a small country dominated by Tokyo
* Overpriced leased lines


What's Needed for Japan's Catch-Up
* More support, especially financial, from government and industry
* Better-educated bureaucrats
* More coordination between existing networks
* Elimination of barriers imposed by government between academics and industry
* More free-access systems
* More applications software
* Faster links
* Lower tariffs
* More researchers
* More links with other Asian countries




WIRED Online Copyright Notice


Copyright 1994 Ventures USA Ltd. All rights reserved.


This article may be redistributed provided that the article and this notice
remain intact. This article may not under any circumstances be resold or
redistributed for compensation of any kind without prior written permission
from Wired Ventures, Ltd.


If you have any questions about these terms, or would like information
about licensing materials from WIRED Online, please contact us via
telephone (+1 (415) 904 0660) or email (info () wired com).


WIRED and WIRED Online are trademarks of Wired Ventures, Ltd.