No. 2 — June 4, 1983

This is the second issue of the newsletter, the purpose of which is to fill the need for better communications in the field of technical translation, and to disseminate useful facts and figures about translation and the information sciences as they relate to Japan, placing special emphasis on the requirements of professional technical translators working between Japanese and English. Issues will be published occasionally, that is, whenever there is anything to communicate. The publication is free and uncopyrighted. Please circulate the newsletter among your friends who are working in this and related fields. Contributions are most welcome.

In this issue you will find information about Japanese database systems (JICST, JAPATIC, NEEDS, etc.), translation industry news (Mitsui Bussan, Nippon Gijutsu Boeki, etc.), notes on dictionaries (the blockbuster dictionary from Inter Press, a trilingual physics dictionary, dictionaries on pharmacology and fine chemicals, reference books on Chinese medicine, directories of information sources and Japanese scientific periodicals, acronyms, a rare War Department dictionary from World War II, and an amusing "Japanese business glossary"), a book review of "Fifth Generation: Artificial Intelligence and Japan's Computer Challenge to the World," and notes about Japanese (and Korean) word processors.

* * *

In the last issue I appealed for information about the Japan Information Center of Science and Technology (JICST) (Nihon Kagaku Gijutsu Joho Senta), which operates a number of databases which are called JOIS, which stands for JICST On-line Information System. Thanks to Derek Freyberg and other readers, I was able to obtain a wealth of information about the JICST and some other databases. I have also noticed a number of recent newspaper articles about the current state of the database business in Japan.

The Ministry of International Trade and Industry (MITI) on May 29 made up a list of the database services operating in Japan and the types of information offered by them. According to MITI, there are 42 businesses and agencies engaged in supplying database services in Japan. They offer 456 different sorts of information, three-fourths of which have to do with natural sciences, technology and business. There is a group called the Deetabeesu Saabisugyoo Renraku Kondankai (abbreviated as DB-kon). It is a business association with a membership of 33 privately operated database services and was established in 1980.

The JICST is a tokushu hojin under the Science and Technology Agency. It has a staff of 326 persons and is the central agency for scientific and technical information in Japan, operating databases containing some 14,750,000 items of scientific and technological information. The data is accessed by some 1,200 terminals throughout the country.

Another important database service is that operated by the Japan Patent Information Center (JAPATIC) (Nihon Tokkyo Joho Senta), a shadan hojin operating under the Patent Office. It has an on-line retrieval which is connected to 770 terminals using public lines and to 57 terminals using dedicated lines. Its services overlap somewhat with the Patent Information Section of the Japan Institute of Invention and Innovation (JIII) (Hatsumei Kyokai), and plans are being made to merge both of these agencies.

It has been announced that the on-line retrieval services of the JICST and the Japan Patent Information Center will be merged so that a user can access both databases simultaneously from a single terminal.

JAPATIC is going to be merged with the Patent Information Section of the Japan Institute of Invention and Innovation (JIII) (Hatsumei Kyokai). The two mutually redundant organizations are to be combined by the end of March, 1985. JAPATIC will be disbanded and will merge with the JIII Patent Information Section to form a new corporation (zaidan hojin).

The eventual merging of JAPATIC and JIII will mark the unification of all government-run patent information services, and the planned combination of the JICST and JAPATIC services is expected to facilitate information retrieval in Japan. Do these merger moves point towards an eventual total unification of the largest databases?

Another frequently mentioned on-line database service is NEEDS, (Nikkei Economic Electronic Databank System), operated by the Nihon Keizai Shimbun. It contains information about economics, companies, and stocks and can be accessed from the U.S. by an international hook-up called VENUS-P.

These database services appear to be utilized mainly by companies and local governments. The Japan Management Association (Nihon Keiei Kyokai) this year made a two-month survey about the state of utilization of databases in Japan and found that half of the large Japanese companies and local governments are using in-house information retrieval systems using computers or microfilm, and more than 30% of them also use external data-base services. Replies from 232 large companies and 35 local governments were tabulated. 53.2% of the entities replied that they were using computers for information retrieval, and another 25.2% indicated that they wished to do so in the future. 54% of them were using microfilm, and 21.2% intended to do so in the future. Only 2% indicated that they were currently using optical disk filing systems, but 48% of the total expressed a desire to use them in the future.

38.4% of the companies surveyed had their own in-house databases with some means of information retrieval. 32.4% of the companies and local governments were utilizing external database services, and 30.8% said that they intended to use them in the future. The three most widely used databases among manufacturing companies were:

  1. JOIS-II, the database system of the JICST
  2. PATOLIS, the system operated by the Japan Patent Information Center
  3. DIALOG, a system operated by Maruzen and Kinokuniya, two Japanese bookstores

The three most widely used databases among non-manufacturing companies were:

  1. QUICK a financial database operated by the Shikyo Joho Senta
  2. NEEDS the system operated by the Nihon Keizai Shimbun

The Japanese government has announced that it intends to promote sales overseas of Japanese databases, but except for the NEEDS database, I have not yet seen any concrete proposals about means by which Japanese-language databases could be accessed from overseas terminals. There seems to be a growing awareness in Japan of the country's "backwardness" in the collection and dissemination of scientific and technological data and of the need to build databases which are usable internationally, if only as a means of assuaging foreign countries increasing dissatisfaction with Japan for being only a consumer of foreign-originated information rather than a supplier of its own information.

* * *


Mitsui Bussan (Mitsui & Co., Ltd.), one of Japan's large trading companies, has decided to establish a company devoted exclusively to the language-training and translation business. An article in the Nikkei sangyo shimbun (May 14, 1983, p. 11) says that the company has established a wholly owned subsidiary company called the Bussan Kenshuu Sentaa which will absorb the Mitsui staff hitherto engaged in these activities.

The staff consists for now of about 10 employees, including foreign specialists in English, Spanish and French. It will carry on the work of foreign-language training of Mitsui employees and translating and editing its in-company publications in foreign languages, but in the future the company expects to expand into the general market and accept orders from a wide range of clients.

This move seems to be part of the recent trend for the Japanese trading companies to move away from goods and to seek profits in service businesses.

* * *

Nippon Gijutsu Boeki Co., Ltd. (NGB) is a company specializing in patent documentation which provides translations of patent and other materials. It was established in 1959 and claims to have a full-time staff of 150 persons as well as a pool of more than 150 translators and abstractors in almost all technical fields. The company offers a full range of patent watching, abstracting and translating services. It has been supplying the English abstracts for the Derwent company for many years. Its address is:

Nippon Gijutsu Boeki Co., Ltd.
Translation Center
Kasumigaseki Bldg., 32F.
1-5, Kasumigaseki 3-chome
Chiyoda-ku, Tokyo 100 Japan

* * *

Another company supplying patent and technical translation and research services is:

Intertec Corporation
Toranomon Akiyama Bldg.
No. 22-13, Toranomon 1-chome, Minato-ku
Tokyo, Japan

* * *

Many thanks to Derek Freyberg for the above information about NBC and Intertec, and also for information about JICST.

* * *


The gigantic two-volume dictionary, Kagaku gijutsu 25-mango daijiten, which I mentioned in the last issue, has been published (as of April 25, 1983) and received here in May. It has an English title "Inter Press Dictionary of Science and Engineering." One volume (1888 pages) is "Japanese-English", and the other (1927 pages) is "English-Japanese." The prepaid price is ¥80,000 ($334.73) until May 31, but after that date the price goes up to 88,000 ($368.20). By a special dispensation, I paid $399 for my volumes which were airlifted to San Francisco by the ever-helpful OCS America. Inter Press is a publishing company which has for years been putting out a little magazine called Kogyo Eigo [Industrial English] and also a line of textbooks for Japanese to learn technical English. It has from time to time published its own series of technical glossaries on specific technical subjects, and the present daijiten appears to be an outgrowth, on a enormously expanded scale, of these earlier glossaries. The entries in the Japanese-English volume are arranged according to the gojuonzu system, which is perhaps less useful than a Romanized arrangement (for example, words beginning with the syllable ha are interfiled with words beginning with syllables ba and pa, and it is sometimes inconvenient to have to plow through dozens, perhaps hundreds, of unrelated words in order to find just the one you are searching for). The phonetic entry is followed by the entry in traditional Japanese kanji-and-kana script. The English equivalent(s) then follow. A unique feature is that each English equivalent is labeled for its source or sources. At least four main sources are used: the Gakujutsu yogo-shu series published by the Ministry of Education; the JIS standards (A0002 to Z9211); computer vocabularies published by IBM and others; and the Inter Press glossaries (labeled IP). The vocabularies from the first two sets of dictionaries are already available elsewhere but have never been combined in a single volume (see the review of the JIS kogyo yogo daijiten in the last issue of this Newsletter). The IBM and IP glossaries have not been widely available in the past. The fact that the dictionary brings together 250,000 words from such a large range of pre-existing sources, identifying exactly the source of each entry and the field in which each equivalent is used, makes it a first-class research tool for every translator, since so many Japanese terms have more than one English equivalent, and the same Japanese word sometimes corresponds to different English equivalents in different technical fields. My impression is that this dictionary more or less supersedes the entire Gakujutsu yogo-shu series, on which we have been relying heavily for so many years, as well as the JIS publications (the aforesaid JIS dictionary is still of use because of its detailed definitions and illustrations). The Inter Press dictionary is a truly stupendous accomplishment. I don't like to exaggerate, but how else can one describe a 3,815-page blockbuster in two hefty volumes, each one weighing I think about 7 pounds? It may be heavy to pick up but it certainly belongs on the desk of every technical translator, no matter how many other dictionaries he or she already has, and I suspect that a very large number, maybe most, of the specialized dictionaries we have been collecting at such great expense for so many years are now superseded forever. This is the dictionary which I would recommend as the first choice for anyone beginning a career in technical translation. Buy this one and you won't have to buy the other twenty or thirty dictionaries which it supersedes. To help you judge for yourself, on the next page I'm giving a sampling of some entries from the J-E and E-J volumes.

げんゆきょうどうゆにゅうきち[原油共同輸入基地]central terminal station [IP・プラント]

げんゆさんち[原油産地]crude source [IP・プラント]

けんゆたんく[検油タンク]inspection tank[F0026・造船]/ observation tank[B0113・燃焼][F0026・造船][IP・プラント]/ test tank[学術・船舶]

げんゆとくりつ[原油得率]yield of crude oil[IP・プラント]

げんゆなまだき[原油生だき]crude oil fueling[IP・エネルギ]/ direct combustion of crude oil[IP・公害]

げんゆのき[原油の基〔石油〕]base of crude oil[学術・化学]

げんゆぼう[原油棒] dipstick[B0110・内熱]oil dipper[B0110・内熱]/ oil-level gauge[B0110・内熱]/ oil-level stick[B0110・内熱]

げんようき[現用〔航空〕幾]active aircraft〈航〉[IP・軍事]

げんようご[現用語]living language [IP・プラント]

gauge-piece アダプタ(あだぷた)[C0201・ヒューズ]

gauge pipe 計圧器官(けいあつきなん)[F0026・造船]

gauge pressure ゲージ圧(げーじあつ)[B0119・水車][B0131・ポンプ][B0132・送・圧][学術・機械][学術・計測][学術・採鉱治金][学術・船舶][学術・電気]/ ゲイジ圧〔力)(げいじあつ(りょく))[B0120・空圧]/ ゲイジ圧力(げいじあつりょく)[学術・計測]

gauge rod 標尺(ひょうしゃく)〈測〉[IP・軍事]

gauge-rod 尺づえ(しゃくづえ)[学術・機械]

gauge-strut ゲージストラット〔鉄道)(げいじすとらっと)[学術・土木]

gauge tester ゲージ試験機(げーじしけんき)[学術・船舶]

gauge tie ゲージタイ(げーじたい)[E3013・鉄道]

gauge-tie ゲージタイ〔鉄道)(げーじたい)[学術・土木]

* * *

Kim Mine: Wa-ei-ro butsurigaku jiten, Japanese-English-Russian physics dictionary ("Yaponsko-anglo-russkii fizicheskii slovar"), Moscow, Russkii Yazyk, 1982. A nicely done new dictionary containing about 24,000 terms. 885 pages. The main text (pp. 11–523) has the Japanese terms arranged alphabetically in the Hepburn system of romanization (slava bogu, not the clumsy Cyrillic transliterations), followed by their Japanese kanji-kana forms and the equivalents of each in both English and Russian. The main part is followed by Russian and English index sections referring back by number to the Japanese terms. Available for $18.95 from Victor Kamkin Inc., 12224 Parklawn Dr., Rockville, MD 20852.

* * *

The Nikkan Kogyo Shimbun announces publication of the 1983 edition of its Sangyo yogo jiten which sells in Japan for ¥1,400 (around $5.85) and contains some 2,600 basic words about industry.

* * *

The Hirokawa Shoten (Hongo 3-27, Bunkyo-ku, Tokyo) has published its one-volume dictionary of pharmacology: Hirokawa Yakkagaku daijiten. It has 1500 pages of main text and a 400-page index of foreign words. The special after-publication price is ¥28,000 (about $117.15) until September 30, after which the price goes to ¥32,000 (about $133.89).

* * *

A company called Shii-emu-shii (CMC?) (Kusumoto Bldg., Uchikanda 1-11-13, Chiyoda-ku, Tokyo 101) has published a dictionary of fine chemicals (Fain kemikaru jiten) It has 1100 articles and 1366 pages and sells for ¥30,000.

* * *

Now for something completely different. I was fascinated by a full-page advertisement in the Nihon keizai shimbun on May 18, 1983 announcing publication of a whole series of Japanese-language reference books about Chinese medicine by a company in Kyoto called Yuukonsha in collaboration with a Chinese company called Jinmin Eisei Shuppansha. The books which are available now include:

Genshoku Chuugoku honzoo zukan [Illustrated compendium of Chinese medicinal plants in color], a set of 4 illustrated volumes devoted to Chinese medicinal herbs. The volumes have some 5,000 colored paintings of the herbs, and each herb is identified by its Chinese name as well as its Latin and Japanese equivalents. There are full descriptions of the place of origin, distribution, processing, chemical components, pharmacology, doses, etc. The four-volume set costs ¥132,000 ($552.30).

Kampoo igaku daijiten [Big dictionary of Chinese medicine), a two-volume set. Volume one is devoted to yakubutsu (drugs). It has 7,000 entries, 2,700 of then with detailed explanations. The articles are written in Japanese and contain Latin and Japanese equivalents of the Chinese names. Volume two is devoted to yakuhoo (prescriptions) and contains more than 9,000 prescriptions which have been mentioned in the literature over a period of some 1,800 years. The two volumes cost ¥60,000 ($251.05).

Chuui rinshoo taikei [System of Chinese clinical medicine), a 6-volume set on clinical medicine. Both Chinese terms and Western terms are used so the books can be used also by Western physicians. The six volumes cost ¥85,000 ($355.65).

The publisher's name and address in Japanese are:

㈱雄渾社 〒606 京都市左京区百万通東20m京都大学前
Tel: 075(722)5151〔代)
Telefax: 075(722)5152

The following two publications are useful:

Directory of information sources in Japan 1980, edited by Japan Special Libraries Association, Senmon Toshokan Kyogikai (Sentokyo). Published by Nichigai Associates, Inc. Distributed by Kinokuniya Book-Store Co., Ltd. Price in Japan ¥10,000 ($41.84).

Directory of Japanese scientific periodicals 1979 (Japanese title: Nihon kagaku gijutsu kankei chikuji kankoobutsu mokuroku) Compiled by National Diet Library. Distributed by Kinokuniya. Supersedes the previous 1974 edition. Price at time of publication was ¥6,300 (that would be $26.36 but the price must be more now). The index is especially useful, since it contains the Japanese titles, English titles, and many of the publishing agencies (universities, learned societies, government departments, prefectures, etc.) all arranged in one alphabet. This directory is useful almost on a daily basis.

* * *

There exists a very rare vintage World War II dictionary of technical Japanese which, although dated and flawed in many ways, is still useful (within reason, of course). Its name is Japanese-English technical terms dictionary (War Department Technical Manual TM 3482). The dictionary was compiled during the above-mentioned war, based chiefly on something called the Saishin Koogyoo Daijiten, a 17-volume Japanese dictionary of technical terms published in Japan in 1939, and it was published by the War Department in November 1946 (printed by the U.S. Government Printing Office in 1947). The dictionary has 1041 pages and contains about 150,000 terms. It would seen to be the grand-daddy of all the J-E technical dictionaries we have today. My copy is a facsimile reprint published in 1954 by the Japanese Ministry of Education and distributed by Maruzen. A bibliographical rarity, it would be very difficult to find a copy anywhere at all, but if you are ever lucky enough to find one in a library, or perhaps can get it on an inter-library loan, it might even be worth-while to xerox all 1041 pages of it. I think of it as a flawed masterpiece. Although hopelessly dated and full of bizarre romanization errors, it still can be quite useful even today. Sometimes one is called upon to translate Japanese texts written years ago which still use this vocabulary. And many of the terms are, of course, still in use. There is a marvelous section at the end (pp. 875–1041) which contains thousands of "non-native KANA compounds," i.e. words like MAIYAA NO SHIYAKU, "Mayer's reagent."

* * *

Acronyms, anyone? There is a Dictionary of acronyms in science & technology published in 1973 by the OHM-Sha. The dictionary has 962 pages, and the editors claim that it contains some 50,000 abbreviations, many of them still in use. Most of the acronyms are English, but some are French, German, Russian, etc. I bought my copy at Kinokuniya in San Francisco for $40.00 but it may now be out-of-print. Unfortunately, many acronyms tend to be novelties, and the ones we encounter in our translation work may have been made up a few months ago, so that a dictionary of 50,000 acronyms published in 1973 might not be as useful as we would like.

* * *

If you are tired of acronyms and long for a chuckle or two, you might like to read "Japanese business glossary" by Mitsubishi Corporation (Toyo Keizai Shinpo Sha, 1983, 220 p; available at Kinokuniya S.F. for $5.95.). The glossary was just published on May 2 and is a best-seller in Japan. It defines and explains, in both English and Japanese, several hundred words and expressions which are peculiar to Japanese business culture. The explanations manage to be both humorous and instructive. An example:


Hiru-andon translates as "a lamp in broad daylight." In broad daylight one cannot tell whether a lamp has been lit or not. From this fact, the word is used to refer to a person whose presence or existence is not regarded as important or to describe a person when it is difficult to determine whether he has value or not. However, it does not mean a person who is slow to react (such a person is called keikoto = fluorescent lamp, which takes time to light up) or is mediocre.

In many cases a Japanese organization operates better when its head is a person who does not stand out but who has a kiremono (sharp and able man) as his futokoro-gatana or chief of staff. In such a case, the top man is a symbol and the deputy or chief of staff holds the responsibility. Often, there are a plural number of deputies and/or staff officers, so it does not happen that power becomes concentrated in one person. Perhaps this perception of the organization which has existed in Japan since the old days has prevented the emergence of a dictator.

Consequently, hiru-andon is not necessarily a deprecating reference and kiremono is not necessarily a word of praise. (pp. 64–66)

I also loved the definition of Nemawashi on pp. 128–130.

* * *

Book Review: "The Fifth Generation"

Was it Norbert Wiener who said that everything in the Universe is either matter, energy or information? Whoever it was, there is the theory that the wealth of nations in the future postindustrial society will consist, not in their land, labor and capital, as was the case in the past, but in their information, knowledge, and intelligence. Knowledge is power (and it will be for sale). The smart, or those who can pay for the knowledge, rather than the wimps, will inherit the future. As the Americans sink complacently into a decline, the Japanese, whose national passion is knowledge, "are moving rapidly and eagerly into a well-educated, information-rich postindustrial society."

At any rate, this is the possibility contemplated in a truly stimulating new book by Edward A. Feigenbaum and Pamela McCorduck called The Fifth Generation (Addison-Wesley Pub. Co., first printing March, 1983). Feigenbaum, I note, is a professor of Computer Science at Stanford University and an authority on artificial intelligence, and McCorduck (author of another book called Machines Who Think is a New York-based writer interested in artificial intelligence. Both of them have visited Japan and have some insightful comments to make about that country.

A clue to the book's thrust is given by the sub-title: "Artificial Intelligence and Japan's Computer Challenge to the World." The mass production of intelligent machines — what the Japanese like to call the Fifth Generation — is, according to the authors, omparable in human intellectual history to the invention of the printing press. The Japanese Fifth Generation project, which was inaugurated on April 14, 1982 with the founding of the Institute for New Generation Computer Technology (ICOT) (Shinsedai Kompyuutaa Gijutsu Kaihatsu Kikoo) is an ambitious, high-risk project structured over a ten-year period. Its object is to make a major effort in the industrialization of artificial intelligence, i.e. to design and produce intelligent computers of a new type ("knowledge information processing systems" or KIPS), and to achieve world leadership in the information processing industry in the 1990's. The project envisions using pictorial as well as voice and written input for these knowledge-based symbolic inference machines. The Fifth Generation project is sponsored by MITI, and the consortium backing it includes eight firms (Fujitsu, Hitachi, NEC, Mitsubishi, Matsushita, Oki, Sharp and Toshiba) as well as NTT's Musashino Laboratories and MITI's Electrotechnical Laboratory. These participants, note the authors, vary considerably in the degrees of enthusiasm and commitment they give to the goals of ICOT, and perhaps after all some skepticism about the audacious project is justified. (You might have already seen the article about ICOT by Bro Uttal, "Here comes Computer Inc.," in Fortune October 4, 1982, pp. 82–90.)

The authors regard the effort as necessary and inevitable. There can be little doubt, they say, about the far-reaching intellectual and sociological implications of intelligent machines. Feigenbaum and McCorduck argue with eloquence that the magnification of the human intellect that intelligent machines represent "is about to bring a qualitative change in human affairs that we can hardly image." (p. 133) In the near future, they assert, no one can plausibly claim intellectual status without an intimate dependence on this new instrument. (p. 212)

For the time being, the competition is world-wide, involving Japan, the U.S., the U.K., France and perhaps a few other nations. The Japanese "are the farthest ahead in perceiving where the new wealth of nations lies," although other nations such as the U.K. and France are also responding to the challenge in their own ways. The Japanese are, say the authors, moving on a grand scale into a field where basic research world-wide has hardly even scratched the surface of the problems. The well-planned, well-organized and well-funded Japanese effort certainly poses a challenge to the American information processing industry, which currently has the lead. The United States, "through a combination of myopia, complacency, and general inertia," may be about to fall behind Japan in this field as it has already done in steels, automobiles and consumer electronics. The danger is not only that this particular industry would suffer a mortal wound; the nation as a whole, warn the authors, could sink to second-rate status, possibly becoming the first great agrarian postindustrial society.

I found in the book few hard facts about MITI's plan for Fifth Generation Computer Systems (which I thought the book would describe in more detail), but I was surprised instead to find a far-ranging, well-reasoned, passionate plea for a revitalization of America. The book's major thrust is in its sixth part, "The American Response," which turns out to be an ambitious analysis of the etiology of the American disease. What is the reason for America's failure? One pathogen seems to be IBM, which displays corporate hostility to the very idea of artificial intelligence and vigorously propagates the idea that computers are "nothing but big, dumb, fast morons." (p. 179) IBM's attitude is characterized by conservatism, caution, and conventionality ("middle-class torpor", p. 183)

But the roots of American inflexibility lie still deeper. For one thing, a coherent industrial policy is lacking. There is no American MITI to finance and coordinate large-scale research, and no ICOT-like central laboratory exists. Funding for education and academic research in computer science is on deplorably low levels. There is a disturbing anti-intellectual streak running through American life. Even the American intellectuals, most of them, haven't the faintest idea of what is going on, or else they regard the fascination with electronic technology as barbaric. "Can a nation that so disdains the life of the mind," ask the authors, "summon up the will to enter, let alone compete in, a world where knowledge is the overriding economic concern?" (p. 210)

The defense implications of artificial intelligence are discussed in an interesting chapter, in which full credit is given to the Department of Defense for supporting research in this field for two decades. It is essential, argue the authors, that we have the best in intelligence. "Our defense industry must obtain and retain a strong position in the new advanced computer technologies." (p. 217) But in the broader sense, national security is a multidimensional state of affairs depending on "healthy, productive industry, agriculture, education, commerce, and government, all thriving on the rapid creation, diffusion, and utilization of knowledge." (pp. 229–230) Revitalization of the industrial base depends on large-scale integration of intelligence and knowledge technology into industrial processes. The authors propose the establishment of a grandiose, NASA-scale national center for knowledge technology.

Be that as it may, the authors have succeeded admirably in making their point. The whole intellectual and social development of mankind is sure to change as a result of universal access to machine intelligence, which is "faster, deeper, better than human intelligence" (p. 236). A knowledge-rich future may correspond to Adam Smith's "utopian" vision of a society of universal plenty which frees the people from subordination and allows true independence of spirit in autonomous action. The authors expect that a technology will soon be in place to permit such a society to exist all over the globe. And it is Japan's goal, according to the authors, to become the world's knowledge-brokers and also to "sell products and services whose design is so knowledge-intensive that their superiority must inevitably claim a large proportion of world markets." (p. 8)

It is unclear to me whether that is really Japan's aim, but many Japanese certainly must be considering seriously what Japan will do when and if it becomes a world center of information. On the other hand, one of the most obvious facts about Japan is its international isolation, and the Japanese would be deluding themselves if they seriously believed that they could become the world's leading knowledge-broker or information center without changing many aspects of their institutional life. As we translators all know, the linguistic problems alone are sufficient to prevent almost all communication as things stand, without giving Japan an additional role to play as the world's knowledge broker.

There have been ruminations lately in Japanese newspapers about this very question. An anonymous columnist in the Nihon keizai shimbun recently (May 11, 1983, p. 15) asked: "Can Japan become the world's information center? We hear that an advanced information society will appear as a result of the new industrial revolution which is now in progress. If Tokyo cannot become the world's information center, then Japan must give up the idea of being the first to reach the goal of an advanced information society."

"It seems to us that information generally converges seeking two elements. One is power military, political, and economic power. In this power we should include commonly shared racial and linguistic traits and the thickness of cross-national personal ties based on them.

"Another element is freedom. Information converges towards places where there are equal opportunities, or in business terms, where there are abundant commercial chances. Even if a powerful political regime exists, when the power is authoritatian the information converging there will tend to be quite distorted."

No doubt the author considers places like New York and London, rather than Peking or Moscow, to fit these definitions of places where information naturally tends to converge. The author applies this reasoning to Japan's case. "As far as economical power is concerned, Japan is remarkable. But its military power and its cross-national personal ties are still feeble." The military lack of power is natural since Japan has consciously chosen not to be a first-rate military power, but the problem is in Japan's international isolation, which the writer attributes Chiefly to the national character of the Japanese and to the peculiarities of their language.

This author suggests, typically, that Japan should compensate for its international isolation by adopting policies of "liberalization." That is, if Japan is accused (from the outside, as a result of some sort of international friction) of being illiberal, then it should compensate for this by adopting measures aimed at "liberalizing" its institutions — allowing more imports of oranges or beef, liberalizing interest rates, relaxing restrictions of one kind or another. The article concludes with some rather incoherent exhortations to the Finance Ministry and the Ministry of Posts and Telecommunication to adopt "liberalization policies."

In the final analysis, there is nothing at all to guarantee that an advanced information society, in Japan or elsewhere, would be any more livable than a backward tribal society, at least unless society clearly recognized liberation of the individual human being from the rigors of social regimentation as an important goal. A profusion of advanced computers alone, without commonly accepted ideas about their use in broadening the intellectual horizons of their users, would not greatly improve the quality of life. At any rate, this is one of the important conclusions which I think can be gathered from this extremely provocative book by Feigenbaum and McCorduck.

* * *


The Japan Business Machinery Manufacturers' Association on May 30 announced its predictions about the future growth of the Japanese business machinery industry. The forecast calls for an average growth of about 12.5% annually in the production of business machinery and other office automation equipment during the next decade. An especially high rate of growth is predicted for Japanese word processors, which, are expected to have an annual growth in production of 47.7% during the next five years. By the year 1993 the scale of production is expected to reach a level about 12.8 times the present level. The Association predicts that the production of Japanese word processors will be worth ¥36 billion ($150,627,615) in 1982, ¥64 billion ($267,782,426) in 1983, ¥450 billion ($1,882,845,188) in 1988, and ¥820 billion ($3,430,962,343) in 1993.

The pioneer in developing Japanese word processors (and also optical character readers) is Ken'ichi Mori, who since April of this year has been director of Toshiba's Information Systems Laboratories. Around 1971, when Mori was engaged in research on speech and pattern recognition at Toshiba, he conceived the idea of producing a Japanese word processor. A prototype word processor was completed in 1976, and the product was commercialized by Toshiba in September, 1978. It was called the JW-10 and sold for ¥6,300,000 ($26,359 at a recent 1983 exchange rate).

Japanese manufacturers shipped a total of 11,000 units of word processors during calendar year 1981. The number increased to 35,000 during 1982, and this is expected to increase further to 70,000–100,000 during calendar year 1983. Lower-priced models have been coming on the market one after another. In May, 1982 Fujitsu marketed its "My OASYS" which sold for ¥750,000 ($3,138) including keyboard, CRT display and printer. NEC and Japan Digital Research both announced similarly priced models in the same month, and later in the year Ricoh and Hitachi also marketed lower priced products. In October, 1982 Toshiba announced its TOSWORD JW-1, which used a liquid crystal display instead of a CRT. It was priced at ¥598,000 ($2,502). The competition continued in 1983. In March of this year, Sharp announced its D-800, which also used a liquid crystal display and cost ¥498,000 ($2,083). Pentel has a "Japanese-language electronic typewriter" called the LETA-CON which sells for ¥468,000 ($1,958) (I'm reproducing newspaper advertisements for the WD-800 and LETA-CON on the last pages of this issue). Now there are said to be nearly 30 companies manufacturing Japanese word processors, including those marketing products on an OEM basis.

The low-cost machines selling for less than ¥500,000 are intended for use at home, but word processors intended for office use still sell for around ¥1,390,000 ($5,815) or more, including printer. While they have been developing the lower-end markets, Japanese manufacturers have also not been neglecting the more sophisticated machines. For example, Sanyo Electric Co. announced on May 6 that it has developed a high-quality Japanese-and-English word processor (SANWORD Nippon SWP-3400) which will go on sale August 1 and will combine word-processing capability with graphics. 20 type styles will be available for Japanese and 15 for English without changing the printer head. A graphics-capable thermal line printer will be used, and the standard price per unit will be ¥2,450,000 ($10,251). Fuji Xerox will begin in June marketing its Japanese word processor (Fuji Xerox 85511 JWP) which will cost ¥1,400,000 ($5,857) for the main unit. A 35 cps printer will be available for an additional ¥400,000 ($1,673), or a 63 cps printer with automatic paper feeder for ¥800,000 ($3,347). The most expensive configuration offered by Fuji Xerox would thus cost $9,204.00. (See below for a price comparison with the astronomically expensive Japanese software to be offered later this year for the Xerox Star in the United States.)

In the meantime, IBM has launched its own full-scale offensive to cut into the Japanese market for office automation equipment. IBM's strategy has been to seek tie-ups with established companies, chiefly Matsushita Electric Industrial Co., as a means to erase its past delays. In March IBM Japan unveiled a small multi-function computer called the IBM Multistation 5550 which can be used as a Japanese-language word processor as well as an online terminal and a personal computer. Production of the unit was entrusted on an OEM basis to Matsushita, printers are supplied by Oki Electric Industry Co., and keyboards are by Alps Electric Co. Prices of this unit range from ¥999,000 to ¥3,050,000 depending on the configuration, and deliveries of the first units were scheduled to begin in June. IBM Japan had originally aimed at selling some 10,000 units in the first year, but during March alone orders for 15,000 units had already been received, and plans were revised upwards to a sales of nearly 50,000 Multistation 5550 units during this year. IBM and Matsushita were forced to increase their monthly production from 700 to over 2,000 units. As a result, production will not be able to keep up with demand, and IBM will be able to deliver only about the 15,000 units for which orders have already been placed; it is expected that this huge backlog of orders will delay further deliveries until some time next year.

Collaboration between IBM Japan and Kanematsu-Gosho, Ltd. has been growing closer, and Kanematsu-Gosho established in April a new company dedicated to sales of the IBM Multistation 5550. In the meanwhile, other companies, including Matsushita itself, are readying their own products to compete with the IBM Multistation 5550. Panafacom, a joint company of the Matsushita and Fujitsu groups, has announced a multifunction 16-bit personal computer series which will be sold by Matsushita under the brand name Operate 7000 and by Panafacom under the brand name PANAFACOM C-280. We can expect to see the Japanese manufacturers turn out a whole new breed of multi-function Japanese-language computers which can be used also as word processors and on-line terminals.

Apart from all the hoopla about IBM et al., something very interesting (at least I think it is interesting) has been going on quietly at Toowa Giken, a small company in Osaka engaged in research and development of office automation equipment. This company announced in May that it has developed a word processor capable of handling both English and Korean and will begin delivering samples this autumn (no price was mentioned). This will be the first time a Japanese company has developed a word processor for a foreign language other than English. The phonetic script (hangul) used in writing Korean consists of 19 consonant and 21 vowel symbols which are combined in a prescribed order into syllable units. One character is one syllable. A syllable can consist of one, two or even three consonants (both initial and final consonants are used) and one vowel. There are more than 4,000 syllable combinations in the language. The Toowa Giken keyboard has keys for the vowels and consonants arranged in a way similar to an English or kana keyboard. The keys for the vowels and consonants are hit consecutively, and the machine automatically forms the symbols into the proper syllabic characters. In addition to the hangul symbols, Korean also uses some 3,000 kanji and some 1,000 special symbols. The word processor also has these in its memory and can create text combining phonetic and kanji symbols in the same way as in a Japanese-language word processor. The company is also developing graphics software for use with its Korean word processor.

Incidentally, the American Xerox company has announced multiple-language software for use with its Star workstation. Among other languages, Xerox will supply Japanese software in the third quarter of this year, but it will be extremely expensive. The multiple-language software costs $1200 without Japanese, and the Japanese software will cost an additional $1200; screen display fonts and a dictionary will cost an additional $800; and Japanese printer fonts will cost $1000. A Japanese-variant keyboard will cost an extra $500. If you add all these up, the total cost of the Japanese package would be $4,700. This assumes that you already have a Star workstation, which costs something like $18,000, unless I am mistaken. Then all you need to get is a laser printer (another $30,000 or so). The bill would be a whopping $52,700. I doubt whether Xerox will sell many of its Japanese-language Star word processors since the most expensive configuration of the Fuji Xerox costs a mere $9,204.

* * *

Let me know if any of you turn up information which might be of interest to professionals working in the field, and I will incorporate it in another Newsletter. June 4, 1983

Donald L. Philippi
715 Tenth Ave.
San Francisco, CA 94118
(415) 752-7735

[Scanned Image No. 1] [Scanned Image No. 2]