A graduate of National Cheng Kung University, and with an M.S. from Purdue University, Wang continued his studies in the United States in 1954, improving his professional engineering status via studies at the Massachusetts Institute of Technology, followed by a Ph.D. in Computer Science at the University of Texas, Austin, on a NASA fellowship. He then accumulated top-of-the-line experience at Honeywell, NASA, and the U.S. Department of Transportation.
Inside a very tidy mini-conference room within Transtech's Taipei office suite—on the 11th floor of a modern business high-rise—54-year-old Wang, looking much younger than his years, recalled his early efforts:
"In the initial stages, I merely worked on a consultant basis. At that time, I was still chief of the computer technology branch of the Transportation Systems Center of the U.S. Department of Transportation. One of my programs was intended to promote Sino-American cooperation in transportation, having to do with Taiwan's export traffic problems—all ports were congested with international liners, waiting anxiously for loading. Mutual negotiations were then being vigorously pursued by Wang Chang-ching, a vice minister of the Ministry of Communications, and his American counterparts. However, this joint effort was suspended as a result of the severance of diplomatic relations between the ROC and the United States."
Wang ended up with spare time, over and above the hours now dedicated to his consulting service, and he spent all of it in the basement of his home, working on a Chinese character computer—his concubine", according to Mrs. Wang. Inspired by Professor Samuel H. Caldwell's research during his study years at MIT, Wang in the early 1970s, began to strain at the possibility of efficient data processing in Chinese characters. Sinoterm—an effective system for composing, displaying, and printing Chinese—was his first brainchild.
Based on Sinoroots, an input method developed and patented by Wang, Sinoterm is a component-entry device utilizing 245 "word-roots"—complete Chinese characters or basic strokes or shapes—to compose Chinese characters. Unlike some component-entry systems, the technique does not require component expressions to generate characters in real time; rather, a sequence of keystrokes is used to generate a disk address, and the indicated character is then retrieved from the attached disk and displayed.
For example, anyone who is familiar with written Chinese knows that the character 傳 can be decomposed into 3 parts, 人 , 車 and, 寸 which make up the correct writing sequence, i.e. 人 finished first, and then 車 , and finally 寸. In typing, 人 is typed first, followed by 車 and 寸 , and finally a "space" key is hit just as in English typing, in order to terminate a word. This concept of decomposition and key stroke sequencing is quite natural and customary for those familiar with written Chinese.
Sinoterm employs a set of 245 word-roots, instead of the 500 previously identified, which enables the use of a shift key technique to reduce the number of operator keys to 128. This makes possible the two-handed operation characteristic of an English typewriter. The combining of word-roots into correct sequence, as previously indicated, follows the identical general rules for writing Chinese characters. The result is a straightforward and practical means of typing in Chinese via keyboard that is easy to learn for anyone who knows how to write Chinese characters. Furthermore, a simple rule of further decomposition is used to eliminate all ambiguities; and within the 245 word-roots selected, there are a set of basic strokes which essentially guarantee the facility to create any new character.
The new keyboard—It sends messages ordering the machine to produce Chinese characters
Sinoterm, in April of 1981, came into its first major application when the 12th National Congress of the Kuomintang, the Republic's governing political party, adopted the technology as a means of efficiently recording the results of an election for membership in its Central Committee. Its merits—ease of operation, time saving, and precision.
Surprisingly, unlike its competitors, Transtech is not enthusiastic at promoting sales of its technologies via ready-made products. "We don't even have a salesman," Wang grinned. "We are a project-oriented corporation. In some ways, we evoke tailors in tailoring shops. We mainly focus on the design and development of whole projects assigned to us by our clients."
Among numerous projects undertaken by Transtech, Wang singled out two others for special comment.
In 1982, Transtech faced its biggest challenge—developing an island-wide Chinese character computer network for the ROC Ministry of Defense (MOD) reserve filing service. Terminals are to be installed in 21 counties or cities and on offshore islands, all connecting with MOD's host computer at Taipei headquarters. Actually, this is the first system of its type, and the most sophisticated such system in the ROC. At present, last-stage acceptance tests are underway. At this writing, the Ministry of National Defense was displaying the system at the Annual Computer Show at the Taipei World Trade Center.
Another significant project, a pace-setter in the development of Chinese character computer science, is the so-called RLG/CJK terminal. In February of 1981, the Research Libraries Group Inc. (RLG), a public research organization consisting of 25 leading U.S. university and public libraries—including Yale, Cornell, Columbia, Library of Congress, etc.—circulated a request for proposals to nearly three dozen firms in the United States, Japan, and the ROC. The RLG was soliciting proposals concerning development of a multi-language terminal (Chinese, Japanese, Korean) to be programmed to emulate RLG's existing intelligent terminals, and to use RLG's network communications protocol—the Research Library Information Network (RLIN). Though eight potential vendors, including prestigious IBM, responded, no proposed system was in a state of "off-the-shelf" readiness for RLG's purposes. After a very thorough investigation, RLG entered into a joint development agreement with Transtech for the development of what has come to be known as the RLG/CJK terminal based on the existing Transtech product, Sinoterm.
In the Spring of 1983, the U.S. Library of Congress, with the largest holdings of East Asian materials in the United States, and far and away the largest producer of original bibliographic data on East Asian materials, took delivery of the first eight of a scheduled total of 24 RLG/CJK terminals, for cataloging online into the RLIN. At the same time the East Asian library of the Hoover Institution at Stanford received its four RLG/CJK terminals, and began cataloging into the RLIN. By the end of 1983, a total of 13 institutions, with 54 RLG/CJK terminals, will be using RLIN for the creation, storage, and retrieval of bibliographic records containing Chinese, Japanese, and Korean character data.
The president of RLG, commenting on the first RLG/CJK terminals, called them "a revolution in East Asian librarianship" in the United States.
The most important element in the project, to Wang, is the adoption of the Chinese Character Code for Information Interchange (CCCII), an international confirmation of the ROC's code device. Thus the exquisite, traditional Chinese characters are guarded from the pressure of a pattern of simplified substitutes promoted by the Chinese Communist regime.
Wang sees the heated competition on technical adaptations to the computer age in terms of historical analogy: "Today's situation evokes the fierce fighting in the Spring and Autumn Period and during the Warring States Period (770-222 B.C.). No one will know what technique will be the winner until the last dust settles. The users must make the ultimate decision. I am, though, extremely confident in my system."
From a small basement workshop to the current firm of 20-odd employees, Wang, who confesses to being a very conservative proprietor, has observed deliberate caution towards expansion: "Instead of a quick pace, I prefer steady growth on a very sound base. I have no plans to become a millionaire. Transferring technology is my sole career goal."
Discussing Transtech's future prospects, however, Wang takes the bit: "We plan to focus on the development of graphics and mapping techniques in connection with Chinese characters. Office automation presents another challenge." In addition to obtaining a lion's share in the library market, Wang also plans to develop Chinese-language teaching aids, and applications for the printing industry and the translation field.
The development of the Chinese character computer is still in the pioneering stage. However, Transtech, with its initial accomplishments, is lighting the way for all those struggling along gold rush road.