Excerpted from The Chinese Computer: A Global History of the Information Age, by Thomas S. Mullaney. Published by The MIT Press. Copyright © 2024 MIT. All rights reserved.
During the early rise of consumer PCs in the 1980s, no Western-designed CPU, printer, monitor, operating system, or programming language was capable of handling Chinese character input or output—not “out of the box,” at least. For the better part of the history of computing, computing technology has been biased in favor of certain alphabetic scripts—none more so than the Latin alphabet. In the 1960s, the development team behind ASCII (the American Standard Code for Information Interchange) determined that a 7-bit coding architecture and its 128 addresses offered sufficient space for all of the letters of the Latin alphabet, along with numerals and key analphabetic symbols and functions. Chinese characters, by comparison, would have demanded no less than 16-bit architecture to handle its more than 60,000 characters. And of course, long ago Western engineers piggy-backed on the preexisting typewriter keyboard, using the two-dimensional Shift key to toggle between lower and uppercase letters (Chinese has no alphabet, of course, nor uppercase or lowercase). Whether in terms of character encoding, computer monitors, programming languages, disc operating systems, input surfaces, optical character recognition algorithms, or otherwise, the early history of computing has, in many ways, been the story of one digital “Chinese exclusion act” after the next.
What ensued during the 1980s was a period of hacking and modification—”modding,” to use a common term of art within engineering circles. Element by element, engineers in China and elsewhere rendered Western-manufactured computing hardware and software compatible with Chinese. It was a messy, decentralized, and brilliant period of experimentation and innovation, all of which made the moment in 2013—as well as China’s current stature in the world of computing and new media—possible.
Commercial dot-matrix printing was yet another arena in which the needs of Chinese character I/O were not accounted for. This is witnessed most clearly in the then-dominant configuration of printer heads—specifically the 9-pin printer heads found in mass-manufactured dot-matrix printers during the 1970s. Using nine pins, these early dot-matrix printers were able to produce low-resolution Latin alphabet bitmaps with just one pass of the printer head. The choice of nine pins, in other words, was “tuned” to the needs of Latin alphabetic script.
These same printer heads were incapable of printing low-resolution Chinese character bitmaps using anything less than two full passes of the printer head, one below the other. Two-pass printing dramatically increased the time needed to print Chinese as compared to English, however, and introduced graphical inaccuracies, whether due to inconsistencies in the advancement of the platen or uneven ink registration (that is, characters with differing ink densities on their upper and lower halves).
Compounding these problems, Chinese characters printed in this way were twice the height of English words. This created comically distorted printouts in which English words appeared austere and economical, while Chinese characters appeared grotesquely oversized. Not only did this waste paper, but it left Chinese-language documents looking something like large-print children’s books. When consumers in the Chinese-Japanese-Korean (CJK) world began to import Western-manufactured dot-matrix printers, then, they faced yet another facet of Latin alphabetic bias.
The politics of early dot-matrix printing—this embedded Latin-alphabet centrism—ran even deeper, as discovered in the early work of Chan-hui Yeh, the developer of the 120-shift IPX device. When Yeh set out to digitize and print Chinese characters based on a bitmap grid of 18-by-22, his initial idea was an obvious one: to shrink the size of existing printer pins so as to pack more of them on the printer head. More pins, with smaller diameters, would in theory solve the problem. But Yeh’s quickly discovered solution would not be so simply carried out.
The Latin alphabetic bias of impact printing, Yeh found, was baked into the very metallurgical properties of printer components themselves. Simply put, the metal alloys used to fabricate printer pins were themselves “tuned” to 9-pin Latin alphabetic printing; reducing the pins’ diameters to the sizes needed would result in deformation or breakage. To phrase this another way, the recipes for the metal alloys used to fabricate printer pins were painstakingly calibrated to the standard that worked for Latin alphabetic printing: 9-pin printer heads with each pin’s diameter measuring 0.34mm. To use an alloy significantly more durable than this would have been a needless expense from the perspective of manufacturers who operated under the (likely unconscious) assumption of “A through Z.”
To compensate, engineers in China, Taiwan, and elsewhere devised a series of workarounds. In one mod, some engineers tricked Western built printers into fitting as many as 18 dots—two passes of a 9-pin printer, that is—within roughly the same amount of vertical space as nine conventionally spaced dots. So rather than being twice as tall as their Latin letter counterparts, Chinese characters would be the same height, but twice as dense in terms of pixel placement.
The technique was as ingenious as it was simple. First, an initial array of dots was laid down during the first pass of the printing head. But then, instead of laying down the second array of dots beneath the first, they reprogrammed the printer into registering this second batch of 9 dots in between the first set—like the teeth of a zipper, fastening together. Specifically, engineers rewrote printer drivers to modify the machine’s paper advance mechanism, refining it so that it rotated at an extremely small interval so as to tuck the second set of 9 dots inside the first set.
Mods such as the ones described here were essential to early 1980s-era Chinese computing. Indeed, they formed a core and accepted part of the early economy of Chinese microcomputing—accepted in some cases even by the foreign companies whose products were being “hacked.” A prime example here is the early history of the Sitong Group, known in English as Stone. Established in May 1984, Stone quickly rose to prominence as one of the most important players in early Chinese electronic typewriters and word processors. Many don’t recall, however, that the company’s first production and marketing strategy was focused squarely on modding—specifically on retrofitting the Japanese-built Brother 2024 printer for the Chinese market.
As the Stone employee Wang Jizhi recalled, there was only one 24-pin printer available on the mainland Chinese market circa 1984: the Toshiba 3070, imported by the Fourth Machinery Ministry. Given the staggering cost of this machine (with an import price of more than US$1,000), Wang and others at Stone saw an opportunity. Wang caught word that the Beijing Computer Technology Research Institute had gotten hold of a Brother 2024 24-pin printer. Although capable of printing Japanese kanji and kana, they learned, Chinese-language output came out garbled and unusable.
Executives at Stone launched their new company effectively on the basis of “copycatting” the Brother 2024, rewriting the printer driver with the goal of rendering it compatible with Chinese. The team got in touch with a researcher at the Chinese Academy of Sciences, Cui Tienan, hiring him and three of his colleagues. Cui and his crew managed to rewrite the printer driver in a mere eight hours, thereby opening the door to Stone. Acquiring the machines from abroad, the company would then release them on the Chinese market “after a little bit of computational opening- up work.” The plan worked, Stone’s retrofitted 24-pin printer becoming a runaway success on the small but growing mainland Chinese market. (Meanwhile, Cui and his colleagues received a total of 200 RMB for their labor.)
The following year was even better. If 1984 was Sitong’s year of (reprogramming) the Brother 2024, 1985 was the year of the ITOH-1570 color printer. This printer, also built in Japan, contained an onboard character generator (Hanka) capable of producing kanji characters. The encoding system depended upon the Japanese-standard JIS encoding, however, and so once again a retrofit was needed. Stone assigned Wang Jizhi with the task of “emulating” (fangzao) the Japanese Kanji Card in order to remake it as a Chinese Character Card.
Unlike reprogramming the Brother 2024 printer driver, however, modding the Japanese-built Kanji Card depended on hardware engineering experience Wang did not possess. Intriguingly, however, after Wang managed to get hold of the card itself, as well as the card’s circuit diagram, he was also provided a copy of the firmware program by the Oki company itself. This Japanese company, in other words, was an active collaborator in the “copycatting” process, eager to help Chinese counterparts help them break into the Chinese-language market.
The gambit made good business sense for Oki. By the time Stone held its one-year anniversary celebration—a gala affair held at the Cultural Palace of Nationalities on May 16, 1985, attended by local government notables—the company was clocking annual sales of 31 million RMB (a three-fold increase from 1984). Instead of going to battle with the Stone copycatters, then, it was only logical to collaborate (and profit) with them instead.
It was not until May 1986—two years into the life of the company, after it made its first small fortune by modding foreign dot-matrix printers— when Stone finally released the product for which it is best remembered today: the Stone MS-2400 Electric Chinese Typewriter (Zhongwen dianzi daziji).