Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A prisoner who worked out how to input Chinese characters into a machine

wired.com

Post reply on HN
Print view View original

31–40 of 77 posts

Re: A prisoner who worked out how to input Chinese characters into a machine

#31

Post by odiroot »

est wrote: post #4

He is just one of the early pioneers of Chinese input methods, his particular method is stroke-based and works on 0-9 number pads, went obsolete decades ago, but surely had an influence over later popular ones like Wubi. Today, input methods were dominated by pronunciation based one like Pinyin. The heuristics based "smart" Pinyin input methods was actually invented by a non-tech folk and eventually hired by a then-l…

Pinyin is really peculiar. It makes it so much easier to learn to outsiders.

I know some overseas Chinese people who (not being raised in China) never learned pinyin and prefer the stroke method. Some of the even prefer drawing the strokes by hand (instead of choosing from the keyboard).

Any Chinese word, I've ever learned, I can only input using pinyin. The stroke method adds so much more mental overhead.

Re: A prisoner who worked out how to input Chinese characters into a machine

#32

Post by anonymousiam »

This is a good story for the modern age. Not so much because of the technical achievements of the protagonist, but more as a warning for the perpetrators/victims of class warfare in the world today.

My father-in-law (now deceased) was similarly branded and nearly died while spending several years in a labor camp. He was forbidden to work upon his release, and banished to the countryside along with his family. Decades later, he received a formal apology from the CCP, but no compensation.

Re: A prisoner who worked out how to input Chinese characters into a machine

#33

Post by thaumasiotes »

odiroot wrote: post #31
est wrote: post #4

He is just one of the early pioneers of Chinese input methods, his particular method is stroke-based and works on 0-9 number pads, went obsolete decades ago, but surely had an influence over later popular ones like Wubi. Today, input methods were dominated by pronunciation based one like Pinyin. The heuristics based "smart" Pinyin input methods was actually invented by a non-tech folk and eventually hired by a then-l…

Pinyin is really peculiar. It makes it so much easier to learn to outsiders. I know some overseas Chinese people who (not being raised in China) never learned pinyin and prefer the stroke method. Some of the even prefer drawing the strokes by hand (instead of choosing from the keyboard). Any Chinese word, I've ever learned, I can only input using pinyin. The stroke method adds so much more mental overhead.

> I know some overseas Chinese people who (not being raised in China) never learned pinyin and prefer the stroke method.

I knew someone who used wubi in preference to pinyin input. She grew up in Guangzhou until age 16, so it wasn't an issue of not learning pinyin. The problem, for her, was that she had no way to know the correct pinyin for a character, since she didn't distinguish between a bunch of sounds that are distinct in pinyin.

> Any Chinese word, I've ever learned, I can only input using pinyin. The stroke method adds so much more mental overhead.

If you were capable of writing by hand, this wouldn't be true.

Re: A prisoner who worked out how to input Chinese characters into a machine

#34

Post by jrochkind1 »

dang wrote: post #25
jrochkind1 wrote: post #22

Earlier quoted context omitted.

Agree with other dissenters here that this was not a "generic ideological tangent", it was specific and relevant to the OP. And i found it interesting. I also understand it's difficult to keep a comment on politics, even when it's not a "generic ideological tangent", from prompting a mess, and respect your inteventions to avoid the messes. But this was not a generic tangent.

Ok, you guys, I'll take the point and back off on this one. Thanks for the feedback!

♥

Now watch it turn into a political flamewar and prove you right haha. I mean, you do generally know what you're doing. It was just the template "generic ideological tangent" that got me going, it wasn't at all generic or a tangent... it still might result in a mess though! We'll see!

Re: A prisoner who worked out how to input Chinese characters into a machine

#35

Post by MiddleEndian »

GauntletWizard wrote: post #27
est wrote: post #4

He is just one of the early pioneers of Chinese input methods, his particular method is stroke-based and works on 0-9 number pads, went obsolete decades ago, but surely had an influence over later popular ones like Wubi. Today, input methods were dominated by pronunciation based one like Pinyin. The heuristics based "smart" Pinyin input methods was actually invented by a non-tech folk and eventually hired by a then-l…

Contact list has become the standard - You've probably not even noticed it, but Android's "Gboard" has contacts permissions, and will use your contact's names and nicknames as well as e-mail addresses as suggestions.

I noticed this (albeit with a different keyboard app) as soon as I switched from Windows Phone to Android. Couldn't stand it really, it would auto-suggest random email addresses and even numbers when I was just trying to type words.

The least-bad Android keyboard I've ever used is the HTC keyboard, since it had a setting to only learn words I explicitly told it to remember.

Re: A prisoner who worked out how to input Chinese characters into a machine

#36

Post by rabbits77 »

jjcc wrote: post #17

Here's the bio of the prisoner: https://en.wikipedia.org/wiki/Zhi_Bingyi

I was wondering how he ended up and the biography on Wikipedia is unclear on how his life may have improved after publishing his results. He published in 1978 and passed away fifteen years later. Other than a note that he joined The Party in 1991 there is no information on his last years.

I am not sure what to make of the information that is given. Does it imply he was treated poorly, even tortured, until 1991 when he relented?

Re: A prisoner who worked out how to input Chinese characters into a machine

#37

Post by spicybright »

btbuildem wrote: post #8

I guess I get why he persisted while imprisoned -- anything to do to keep the mind busy. But why would he continue to work afterwards, and benefit the regime that betrayed him? I'd want to watch it burn and then sink, be left as far behind as possible, not enable them to keep up with technological advancements.

This is an interesting question that shouldn't be down voted like it's being.

From how the article is written (although admittedly I only skimmed it), it seems imprisonment didn't have a huge effect on his motivation to solve the problem.

While I can't say I'd be as chipper after being jailed for a year, I can certainly understand the allure of attacking a hard problem with tenacity.

He may have also separated the party from the people he interacted with everyday. In fact, I'd imagine being imprisoned built a lot of comradory with his fellow inmates.

(Would love for someone else more knowledgeable to chime in, I just wanted to start the conversation)

Re: A prisoner who worked out how to input Chinese characters into a machine

#38

Post by VygmraMGVl »

pseingatl wrote: post #23
karmakaze wrote: post #10

It seems like the Hangul[0]/Korean writing system was retrofitted into Chinese with the challenge being to make the imprecise precise. Hangul was designed with the purpose of solving illiteracy and so the character forms were built simply from sound components. Doing it in reverse with a larger unstructured set of symbols is quite the feat. [0] https://en.wikipedia.org/wiki/Hangul

Korean is not a tonal language. But Hangul is great, I wish Thailand would adopt it.

When Hangul was invented, Korean was tonal and Hangul accounted for this with diacritics.

Apparently some dialects of Korean are still tonal.

https://autolingual.com/korean-tones/

Re: A prisoner who worked out how to input Chinese characters into a machine

#39

Post by karmakaze »

pseingatl wrote: post #23
karmakaze wrote: post #10

It seems like the Hangul[0]/Korean writing system was retrofitted into Chinese with the challenge being to make the imprecise precise. Hangul was designed with the purpose of solving illiteracy and so the character forms were built simply from sound components. Doing it in reverse with a larger unstructured set of symbols is quite the feat. [0] https://en.wikipedia.org/wiki/Hangul

Korean is not a tonal language. But Hangul is great, I wish Thailand would adopt it.

Thai spoken language is the opposite extreme of German, vowels are the salient parts and consonants are often barely present and highly interchangeable. I don't know how the Thai written language works or how complex it is in comparison but I imagine it also wouldn't be a straightforward mapping. Quick lookup says it's tonal and analytic [idk what this is] like Chinese and Vietnamese.

Re: A prisoner who worked out how to input Chinese characters into a machine

#40

Post by bllguo »

btbuildem wrote: post #8

I guess I get why he persisted while imprisoned -- anything to do to keep the mind busy. But why would he continue to work afterwards, and benefit the regime that betrayed him? I'd want to watch it burn and then sink, be left as far behind as possible, not enable them to keep up with technological advancements.

is Mandarin the govt's language or his and the people's language? "benefiting the regime" was a mere side effect of his work (in fact the Maoists that persecuted him crumbled in the 70s, so ultimately they did not benefit at all! karmic)

or maybe he just had an intellectual interest in the problem and wanted to solve it no matter what. who knows, but the world is better off for it

Post reply on HN