[LINK] Chinese Computer Keyboards

Stephen Loosley StephenLoosley at outlook.com
Tue Feb 9 01:59:03 AEDT 2021


We are so used to news stories about the technological rivalry of China and the West and intellectual property theft, we often miss out that technologies usually evolve and complement each other in ways beneficial to all sides.

Let me just relate one such history that may be obscure to many people but whose impact affects practically everyone nowadays: computer input. If today, people can type just a few letters, in English, Spanish, French or what not, before your WhatsApp or Google guesses the whole word for you, that technology is partly inspired by the early work of linguists and engineers who had to input Chinese characters by working around the almost universal but oft-criticised English-based Qwerty keyboard.

This intriguing history can be found in a chapter titled “Typing is dead”, by Stanford University historian Thomas Mullaney, who co-edited the new book Your Computer Is on Fire.

The world may be divided into two types of computer users: those who use Latin alphabet and others who have to input non-Western or more specifically, non-Latin scripts. They all, however, have to put up with Qwerty and get round it.

The Qwerty, invented in the 1870s, has often been criticised for being hard to learn, inefficient to use and irrational in its arrangement. Still, it came to dominate first typewriting, and then computing. Indeed, a whole economic theory of inefficiency was first proposed in the mid-1980s by US economist Paul David, who tried to explain why Qwerty, given its clear inferiority to other available alternatives, came to take over the world. He put its survival down to “path dependency”, or what has occurred in the past persists because of resistance to change.

The commercial success of the Remington typewriter ensured Qwerty would become the dominant design. Non-Westerners around the world wanted to use the Remington, too, but Qwerty defied the way they wrote. How did they initially get around that?

“The answer is,” wrote Mullaney, “when creating typewriters for those orthographies with which Qwerty was incompatible, engineers effectively performed invasive surgery on these orthographies – breaking bones, removing parts, and reordering pieces – to render these writing systems compatible with Qwerty.”

And so, early Thai typewriters excluded letters of the Siamese alphabet that didn’t fit on a standard Remington.
In Korea, people experimented with lopping off the bottom half of Korean glyphs or sticking them on the right side. The Ottomans and other Arabs proposed writing in separated letters, when Arabic had always been written in cursive. Mao Zedong and the Chinese communists might or might not have been thinking about the typewriter, but at one point, they wanted to Romanise the Chinese language.

Interestingly, Mullaney points out, it wasn’t just Westerners who imposed the Qwerty on hapless natives. Progressive and educated local elites were often the most enthusiastic.

Such struggles over the Qwerty happened with dozens of non-Latin-based languages and cultures. Mullaney never wrote the phrase “Western imperialism”, but he was clearly describing an obscure if unmistakable aspect of it.

But something interesting happened with Chinese, in which Mullaney specialises. In 1947, Fujian-born Lin Yutang, the great writer, linguist and inventor who died in Hong Kong in 1976, designed a Chinese typewriter with a keyboard that looked like the Remington but used a completely different “input logic”.

A key feature of Qwerty is “auto-advancing”: when you press a key, it creates an impression on the page, and then moves forward one space.

Because of the constraints of writing Chinese character strokes, that obviously wouldn’t work for the language. Instead, Lin’s design, called the MingKwai, allows you to input up to three strokes without making an impression on the page. From those few strokes, fully formed Chinese characters would appear on the top of the machine for you to choose to impress on the page. It’s not clear what the machine’s degree of accuracy was.

Instead of a typing machine, wrote Mullaney, Lin had created “a mechanical Chinese character retrieval system”. His invention was a commercial failure but its underlying input logic would continue to inspire: criteria, candidacy, confirmation. That’s still the same input logic when you send a Chinese email on your Xiaomi smartphone or write a Chinese essay in Microsoft Word. And it’s user-friendly.

The next breakthrough came from MIT engineering professor Samuel Hawks Caldwell and Harvard linguist Yang Lien-sheng. Caldwell didn’t know a word of Chinese but was a specialist in logical circuit design for analogue computers. Then he was intrigued that the Chinese language had “spelling” or rather what we would call written character strokes in a specific order. Yang helped him analyse more than 2,000 commonly used Chinese characters and break their components or strokes down to 22 combinations to fit nicely on a Qwerty-style typewriter. Statistically, they found that the median minimum to specify 2,121 Chinese characters was between five and six strokes.

All that became the basis of contemporary Chinese computing. Lin, Yang and Caldwell had stumbled on what we today call “autocomplete”, which only recently became common in English word-processing but has been the mainstay of Chinese computing since the 1950s.

“The conceptual and technical framework that Lin and Caldwell laid down would remain foundational for Chinese computing into the present day,” Mullaney wrote.

“Every computer user in China is a ‘retrieval writer’ – a ‘search writer.’ In China, ‘typing’ has been dead for decades.”

The overcoming of the Qwerty limitations on input by different non-Latin language groups probably all warrant an intriguing history, but each would require language experts of their own to tell.

What they are all likely to say is that what started off as Western domination ended up as technological liberation for all.


More information about the Link mailing list