[registrars] Surfing the web on a cell phone in Japan, an explosive new phenomenon.
In a prior posting to prepare members who plan to attend ICANN and/or
INET2000 meetings in Yokohama, I mentioned the iMode cell phones -- cell
phones which can surf the web. While the penetration of PCs in Japan is
much lower than in the U.S., the cell phone is greatly expanding the
audience for the Web.
Here is an article from the July issue of The ACCJ Journal (Journal of the
American Chamber of Commerce in Japan.) The hard copy arrived two days
ago. However, it has not been posted on the ACCJ web site yet. The editor
kindly provided me with this copy. There may have been some final editing,
but should be very much like the final version.
Other articles in this issue may be of interest to many of you in this issue:
Inventory Slasher: An Internet start-up builds an auction Web site to help
reduce one of corporate Japan's biggest expenses: mountains of unsold goods.
Investing in Cyberspace (SPECIAL FEATURE): Beware of the pyramids of
cyberspace! Where does the money come from to fuel fast-takeoff Internet
firms, and how does it find a home in Japan.
Advertising Web-Wise: Overall, Japan's advertising industry appears
stagnant, but the Internet brings unprecedented opportunity.
Japanese Code Words: Layers of meaning unfold behind the elements of a
language that for centuries was deliberately used as a barrier against
ACCJ has provided us with several extra copies of this issue for ICANN and
ISOC2000 participants. If you'd like one, corner me at one of the meetings
in the next two weeks.
Deconstructing 'Phone Culture'
How Japan Became a Leader in Mobile Internet
By Ray Tsuchiyama
One can easily pick out the tall American and European visitors wandering
around Hachiko, the worn statue of the world's most loyal dog, in Shibuya,
the youthful, trendy district of Tokyo. In the land of 'Phone Culture,'
like Margaret Mead researching Samoan villagers, they peer at teenagers
holding cellular handsets like talismans. This is such a change from the
1990s—the era of 'Japan-passing.' In wireless technology Japan is two
years ahead of Europe, and Europe is two years ahead of the United States.
How and why did this happen? The answers are complex.
In essence, the Japanese wireless industry is kick-starting the Japanese
economy (NTT DoCoMo, the leading Japanese cellular carrier, has a market
capitalization of over $350 billion, larger than several Asian countries'
GDP combined). As late as 1994, barely six years ago, the Japanese
cellular phone sector was moribund. In spite of establishing the world's
first analog cellular network in 1979, Japan was stuck in a rut. The
Ministry of Posts and Telecommunications kept stubbornly to a leasing
scheme and high tariffs for cellular phones that restricted subscribers to
executives on expense accounts. By the spring of 1994, 15 years after the
first wireless handsets were introduced in Japan, there were only 2.1
At that opportune moment, the U.S. government intervened with what is now
looked upon as a shining example of successful gaiatsu, or outside
lobbying pressure. What emerged was the 1994 Cellular Telephone Agreement,
which, according to Making Trade Talks Work: Lessons from Recent History
(ACCJ, 1997), "abolished restrictions and delays in establishing the
technical networks necessary to meet growing demand for cellular service."
The ACCJ study pointed out that "a further important change was that
customer-owned and maintained cell phones were permitted, removing the
requirement that all cellular telephones had to be leased. (Ultimately)
this brought about new competition, reduced prices and a tenfold increase
in cellular service in two years."
In 1994, the U.S. government was the champion of telecommunications
deregulation and innovative technology, and officials still maintain that
the Japanese consumer benefited from this gaiatsu. In 2000, ironically,
the U.S. is not a wireless world leader, but instead has five incompatible
wireless standards. U.S. carriers are selling American consumers bulky
phones that teenagers in Shibuya would find ridiculously old-fashioned. It
is impossible to name a U.S. carrier establishing a new R&D center in
mobile Internet. To be sure, though, Americans extol themselves as the
leaders in high tech (name a Japanese Microsoft, a Japanese Intel).
When Americans talk about the Internet, they assume there is one model of
Internet access: a personal computer, an Internet service provider (ISP)
account and an all-you-can-use carrier-calling program (free local calls,
taken for granted in the U.S., are unknown to Japan and much of Europe).
The PC and the Internet exploded into U.S. society without government
guidance. No special subsidies existed for Microsoft or the IBM PC
division. Rather, the PC and Internet boom came about by a unique U.S.
combination of excellent research universities, government funding of
basic research and a deregulated flow of research ideas into the private
sector. For example, no one at the Defense Advanced Research Projects
Agency really imagined that research in communications would result in
on-line auctions and cheap air tickets.
Based on this success-because-of-lack-of-government-intervention idea, the
U.S. government took a similar hands-off policy, to 'let the market
decide' in the telecommunications sector. As a result, wireless
subscribers grew substantially in the U.S. However, another result is a
patchwork of warring fiefdoms of different wireless standards. Many
carriers wished to derive as much revenue as possible from their
infrastructure investments for the longest time possible, hence a
confusing mixture of analog, analog/digital and digital networks.
Meanwhile, in the late 1980s, when the rest of the world was still
grappling with the older analog wireless standard, Europe took a tack that
was opposite to that of the U.S. Europe had a legacy of government
intervention in the economy. The wireless industry, still in its infancy,
was to be no exception. A nightmare was that each European country might
go with its own proprietary network, which would be incompatible with
those of neighboring countries. In the worse-case scenario, a Dutch
traveling salesman just a few hours' drive from the German, French, Danish
and Belgian borders, would have to maintain one cell phone for home, plus
four additional handsets that operated according to different wireless
The European Telecommunications Standards Institute (ETSI) dragged
European countries together to create GSM (Global System for Mobile
telecommunications), a digital network based on TDMA (Time-Divisional
Multiple Access protocol). ETSI officials probably never dreamed that GSM
networks would proliferate in a decade beyond Europe into Africa, Middle
East, Asia, and the largest potential market of all—China.
With GSM carriers signing 'roaming' agreements, a Hong Kong businessman
can now carry his GSM handset to Germany, and when a friend in South
Africa calls his Hong Kong number, it rings in a crowded conference hall
in Hanover Messe. More significantly, GSM is a digital network, which
means that far more callers have been added to the subscriber rolls. Also,
smaller and longer-talk-time handsets are available and—the most
far-reaching advantage over analog networks—data, not only voice, can be
transmitted in electronic 'bits,' like computer code. GSM proponents now
claim that more countries have GSM networks than have McDonald's fast-food
restaurants. If true, this is quite a feat.
Telecommunications firms of countries that swiftly deregulated their
telecommunications markets, especially Nordic countries like Sweden and
Finland, had to adjust quickly to competition. In stark Darwinian terms,
the decision to move to GSM was a godsend for firms like Ericsson and
Nokia, outfits dozing quietly near the Arctic Circle that reinvented
themselves as global technological and marketing organizations. Both firms
followed the GSM flag wherever it was unfurled, achieving success in
Europe and then Asia.
With PC standards, the market alone decides the winner. In wireless,
governments play a stronger role, as the allocation of radio frequencies
is linked to national security. Basic choices can spell success or failure
for equipment manufacturers. In the early 1990s, Japan decided against
GSM. Instead it adopted the Japan-only PDC
(Personal Digital Cellular—a variant of TDMA) standard developed by NTT.
(The U.S., as part of its open-market policy, would add GSM to its
wireless-standards basket.) Europeans claimed that Japan excluded GSM so
as to nurture its own telecommunications firms. The Japanese complained
that the European club shut them out (even now, Japanese manufacturers
have barely 10% of the European GSM-handset market). Some claim, though,
that if Japan had adopted GSM and deregulated its telecommunications
market faster, Japanese manufacturers would have had to adapt quickly and,
like the Nordic firms, could have achieved greater success. Even Sony
President Nobuyuki Idei has said that without greater competition in the
Japanese telecom market, "Japan would have a warped computer and
Statistics indicate that Japan lags three years behind the U.S. in PC use
and two-and-a-half years in Internet use. By early 2000 the U.S. had over
50% PC penetration, while Japan was under 40%. The U.S. figure was over
30% for Internet penetration, while Japan was around 20%. On the other
hand, this spring, the number of mobile-phone subscribers in Japan
exceeded fixed-line-phone subscribers. The former topped 57 million, with
the latter stagnant at 56 million. The U.S. has 75 million cellular-phone
subscribers, but Japan leads on a per capita basis.
About 20 million Japanese, or 30% of all Japanese cellular-phone users,
are now using the Internet (compared with about five million in 1997).
Most access the Internet at the office. Also, most of them are male—an
important fact that will figure in a later discussion about mobile
Internet. Only 20% of local schools are linked to the Internet, compared
to nearly 90% in the U.S., so there is a very large untapped market of
young people. Because of low PC numbers, it is no surprise that e-commerce
comes to about $800 million a year in Japan, against $36 billion a year in
the U.S. Fewer than two million Web sites are up in Japan, compared to
more than 30 million in the U.S.
In early 1995, there were fewer than 100 Internet service providers (ISPs)
in Japan. Last year there were nearly 3,500. The top fifteen ISPs occupy
half of the market. The major difference with the U.S. is that, reflecting
the traditional Japanese large-company paradigm, major ISPs are
subsidiaries of large electronics manufacturers, not independent
companies. The top two ISPs are offshoots of Fujitsu and NEC. Matsushita
and Sony also have Internet-subscriber businesses. NTT could not align
itself with one firm; to link-up with Fujitsu was to alienate NEC and
3,000 other ISPs, so there was no choice in the ISP marketplace—this
explains the NTT decision to create its own wireless Internet portal.
As a result, though Internet use is increasing dramatically, Japan is
still well behind the U.S. Unexpectedly, the Japanese wireless market has
raced ahead. With success came unexpected dangers. In 1997, frightening
scenarios were aired about the collapse of the Japan-only PDC digital
network by 2001. Too many subscribers were making calls simultaneously,
which meant that available 'time slots' (hence, the 'time-divisional' in
the TDMA protocol name) fell short of demand. To expand capacity in the
mid-1990s NTT DoCoMo had to spend over $6 billion in infrastructure costs
switching centers, software). Even after that buildup, by the late 1990s
NTT had more consumer complaints about voice quality and dropped calls.
Advertisements by DDI and IDO for new CDMA-1 (Code Divisional Multiple
Access, developed by San Diego-based Qualcomm from military
spectrum-hopping research) handsets stressed better voice quality.
For one scary period, as NTT DoCoMo lost market share, it faced the
dilemma of having to borrow even further to expand an infrastructure based
on decade-old PDC technology. The entire network would be obsolete before
the debts were paid off. The company then noticed that Japanese cell-phone
users were changing handsets at a phenomenal rate—every eight months on
average. It embarked on a program to entice subscribers to use more data,
rather than voice, on its existing network. The objective was to buy time,
while postponing investment in a completely new infrastructure.
Thus, the stage was set for a 'killer app.' The i-Mode project became
Japan's mobile Internet phenomenon. Launched in
February 1999, the first i-Mode model was the 501i series manufactured by
Fujitsu, Mitsubishi Electric, NEC and Matsushita. The
second-generation 502i released this spring has color screens, MIDI
(Musical Instrument Digital Interface, a communications protocol that
allows electronic musical instruments to create computer synthesized
music), music jingles (amazingly, books on cell phone melody tunes in
Japan sold over five million copies last year), and other features.
Technically speaking, i-Mode still employs the PDC network for voice
calls, and a new packet-based data network was built alongside the
existing PDC network. Moving to a packet-based architecture offers
advantages over switch-based hierarchical networks for both data and voice
traffic. With multiple users sharing a single channel, for example,
packet-data services will cost much less than voice services. For the
small-screen i-Mode Web pages, DoCoMo used the proven subset of HTML 3.0,
with some additional tags. For content providers, there was no need to
learn a new language to create Web pages, just some re-configuring for a
As for the success of i-Mode, from zero in February 1999 there were five
million subscribers in March of this year. By 2002 there should be over
ten million i-Mode subscribers—about 30% of all cell-phone users in Japan
accessing the Internet. All new cell phones sold by DoCoMo are now i-Mode.
What is clever about i-Mode is that the 'i' never referred directly to the
Internet. It was about 'I', or 'Me', and the obsession of Japanese youth
with their own fashions and values.
Some observers note that if i-Mode were defined as an ISP, its five
million users would outnumber those of @nifty and Biglobe, the two top
Japanese ISPs, combined. This is not entirely correct. World Web/Internet
pioneers Tim Berners-Lee and Vin Cerf envisioned the Internet as open and
unrestricted to all Web pages. i-Mode is a pass-through; it brokers
content providers to the user, charging fees. It makes creation of content
very easy but is not responsible for content.
i-Mode users connect to the Net at 9.6 kilobits per second. This may sound
laughably slow to an American used to cable modems, but, utilizing packet
transmissions, it's enough. It takes just a few seconds for all the data
to assemble in the handset. Unlike PCs, laptops and PDAs, the i-Mode
handset is connected to the Internet 24 hours a day.
As of last spring, there were 400 application-alliance partners, up from
67 at service launch. Also, there were over 5,000 voluntary
i-Mode Internet Web sites, including seven search engines. Among the
alliance partners are major banks (money transfers, account balances),
securities firms (stock quotes, buy/sell stocks), credit-card companies,
life insurers and airlines (ANA, JAL). There is information on travel
(railway/subway timetables), concerts, news headlines (Asahi, Sankei,
Nikkei, NHK TV), horse racing, restaurant, recipes (Ajinomoto) and on-line
CD and book sales. Entertainment includes network games and even karaoke
Customers are not billed per unit of time. Instead, the fee is about ¥0.03
(approximately three hundredths of a cent) for each packet of data sent
and received. A weather forecast that arrives at the cell phone in five
seconds costs about ¥5. Checking a bank-account
balance costs about ¥10, and transferring funds is some ¥40. The average
total i-Mode monthly bill is about ¥12,000. To many
Japanese, mobile Internet seems very cost-effective compared to PC
Net-surfing at home.
In analyzing i-Mode growth, the only historical model may be PCs, although
this is barely two decades old. Just as PCs were used at first for games,
according to NTT DoCoMo, currently 35% of
use is for entertainment access, while 8% is for tickets, 7% for
travel, 7% for financial information and services, and 22% for
Another correlation between early PC applications and i-Mode is
personalized screen savers. Toy/entertainment company Bandai Co. has a
Hello Kitty Character down-load service with an enrollment fee of ¥100 per
month. This does not seem like much money, but over 700,000 subscribers
(as of last spring) means ¥70 million monthly in gross revenues—a
phenomenal return on investment for Bandai.
The transition to more serious business applications will probably start
with Java i-Mode phones that will be available this fall. Applets will be
not only for interactive games, but also for businesses to use i-Mode as a
link to corporate intranet servers—for scheduling, planning, sales,
logistics, etc. As the PC was transformed into a tool for enterprise
resource planning (ERP) and database management, so also will the
mobile-Internet phone be transformed into a business device. DAT Japan,
Ltd., a delivery company that keeps track of delivery staff and delivery
times via i-Mode phones and a logistics Web page, is already
well-advanced. A case study showed net savings of 90% in
phone-and-operator costs for the company.
To quote Australian-Japanese consumer-marketing expert George Fields: "In
marketing, culture is always first." This fact is essential to the success
of i-Mode. Japanese do not own large homes. They have long commutes. They
gather in urban districts with a plethora of choices regarding eating,
drinking and listening to music. Even while standing in line for a movie
or concert, one has time to explore and choose via i-Mode. Life outside
the home (at least in Tokyo and Osaka) is richer and more vibrant. The
U.S. is definitely a wired world, and the home is a much more convenient,
Almost 70% of i-Mode users are men who have surfed the Net at the office.
Half of all users are in their 20s, and 30% are in their 30s. This profile
resembles the 'early adopter' of the PC era, as Japanese men born in the
late 1970s grew up with Gameboy and Tamagotchi. They are also devotees of
convenience stores, where they pay phone and gas bills and buy concert
tickets. The i-Mode handset can be seen as an extension of the convenience
Carrying an i-Mode phone brings the user much closer to advertisers. This
has an impact, since no company can ignore the advertising potential of
the mobile Internet. The phone has a huge advantage over the PC, which
sits quietly at the home or office until it is turned on.
For greater usage, i-Mode has not yet reached out to the large numbers of
Japanese women who now use the Internet. In order for i-Mode to really
succeed on a mass scale, there must be sites on the Web that will attract
women to i-Mode.
Also, despite its advanced features, i-Mode is still not utilizing natural
Japanese-language text input. The Pocket-Bell input method (just hiragana,
not kanji) and multi-tap (tapping keys to select kanji) are inadequate for
natural-language questions or messages. With natural-language input on a
small keyboard, a phone can really open up the world.
Other Japanese carriers besides NTT are not standing idly by. The No. 3
wireless carrier, Japan Telecom—owned by Japan Railways Group, AT&T and
British Telecom—has been at the forefront of innovation: its brand-name
J-Phone service was the first in Japan to provide SMS (Short Messaging
Services) and Sky Walker is the biggest SMS provider here. J-Phone offers
faster data access since it uses a circuit switch instead of a packet
switch; it has introduced color-screen handsets before DoCoMo. A J-Phone
hallmark feature has been user-selected jingles called 'Sky Melody.'
i-Mode is a preview, not an end in itself. Introduction of the
third-generation network is slated for next spring. NTT DoCoMo is spending
¥800 billion for IMT-2000 service infrastructure. The new standard,
sanctioned by the International Telecommunications Union, allows
high-speed data transmission (from 384 kilobits to 2 megabits per second)
for audio, video and computer-data applications.
The new DDI-IDO-KDD group plans to launch its 3G service by the end of
2002. The DDI group selected the W-CDMA (Wide-Band Code Division Multiple
Access) standard, the same as NTT. This standard was co-developed by NTT
DoCoMo, Ericsson and Nokia, and is an advanced version of a spread-
spectrum standard originally developed by Qualcomm in the U.S. Japan
Telecom is expected to launch 3G service in Tokyo, Osaka and Nagoya in
autumn 2001, with nationwide service in 2003. With partners Vodafone
Airtouch and British Telecom, Japan Telecom selected Nokia as its partner
to design and install all the infrastructure.
Currently, about 15% of all wireless subscribers in Japan are
Internet-accessing subscribers. In five years, perhaps half of all devices
accessing the Internet will be wireless. Europe, reaping the rewards of an
interventionist government strategy, will be close behind. Mobile-Internet
platforms in Europe include WAP (Wireless Application Protocol) and
increasingly transnational networks.
Finally, there's an ironic twist. TRON—the operating system touted by the
Ministry of International Trade and Industry in the 1980s as the means to
reassert Japanese influence over the world PC industry, only to be laughed
off the stage—has been reincarnated as embedded software in some i-Mode
Ray Tsuchiyama, former manager of the Japan Office for the Massachusetts
Institute of Technology, is now Japan country manager for Tegic
Communications, a wholly owned subsidiary of America Online, Inc.