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

51–60 of 77 posts

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

#51

Post by xxfye »

rabbits77 wrote: post #36
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 h…

No post body was provided.

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

#52

Post by xxfye »

1986 wrote: post #43
hbarka wrote: post #24

This man’s biography would make an epic movie. From China, how did he manage to attend and graduate from Leipzig University in Germany during the height of WW2?

Wouldn't this have been during Japanese occupation of China, then? Probably enabled good relations with German universities.

https://en.wikipedia.org/wiki/Sino-German_cooperation_(1926%...

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

#54

Post by 361994752 »

rabbits77 wrote: post #36
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 h…

From the Baike page https://baike.baidu.com/item/%E6%94%AF%E7%A7%89%E5%BD%9D you can see he got a lot of reward during the year 1978-1993, including 院士 (member of CAS) which is almost the highest praise of a scientist. Also he is the leader of many official organizations during those years.

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

#55

Post by WaxedChewbacca »

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.

> But Hangul is great, I wish Thailand would adopt it.

What a weird wish. Does Hangul even have ways to write the in-between consonants in Thai like dt or bp? Does Hangul even have the Thai vowel sound ื (which certainly has no unambiguous way to be written in English)? And then what about tones? Tones in Thai are not written explicitly, but computed based on a number of factors including character classes, length of vowel sound (does Hangul have a way to indicate short or long vowels?), end-of-syllable sound, and tone mark. If you wrote Thai in Hangul, how would you pay homage to loan words from Pali/Sanskrit? Again, just... what a weird wish!

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

#56

Post by hnfong »

thaumasiotes wrote: post #33
odiroot wrote: post #31

Earlier quoted context omitted.

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…

Pinyin is the romanization method for Mandarin (aka Putonghua). From your description it sounds like your friend being more comfortable in Cantonese than Mandarin. (Is there really a difference between not fully understanding pinyin and not distinguishing the different sounds?) Anyway these days Cantonese input methods are getting better (or so I heard, my fingers have muscle memory of Changjie by now). See https://jyutping.org/en/keyboard/

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

#57

Post by seanmcdirmid »

nojs wrote: post #41
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…

As a non-chinese pinyin input is so much easier. One of the cool things about it is you can type a whole sentence just using the first letter of each character. Eg “wxzdnwsmbq” on my phone is sufficient to produce “我想知道你为什么不去”. In English I’d have to type the whole sentence “I want to know why you didn’t go”. The other great thing about it is if you remember the pronunciation of a word but forget the tone and/or char…

Isn’t that sort of like letting auto correct pick the sentences you are texting? One can definitely do that, it’s a cool novelty for Reddit posts, but it isn’t useful.

The nicer thing is they you can type timeless pinyin and then pick the characters you actually meant, useful if you can sort of recognize what you want to say.

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

#58

Post by seanmcdirmid »

eloisius wrote: post #48
xster wrote: post #14

Earlier quoted context omitted.

Taiwan's Zhuyin is 90+% the same as Pinyin except the input is composed with Chinese script radicals (like hiragana) rather than Latin script. > Both generally use Traditional instead of Simplified That doesn't impact the human side of the IME. It's just different output for the same input. And the input (Mandarin pronunciation) is the same for mainland and Taiwan.

Zhuyin aren’t Chinese radicals. They are derived from characters. For example the “b” sounds ㄅ comes from the character 包 bāo.

I thought Zhuyin was like kana because it was designed during the Japanese occupation?

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

#59

Post by est »

hangonhn wrote: post #30
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…

The problem with pinyin is that it's phonetic based and it's based on Mandarin. For large sub-populations of Chinese speakers who primarily use other dialects, it's not as intuitive -- although this is changing overtime as Mandarin becomes more dominant across China. A stroke/character based system preserves a key characteristic of Chinese, which is that speakers of different Chinese dialects can all use the same wri…

> For large sub-populations of Chinese speakers who primarily use other dialects, it's not as intuitive

It's very different than alphabetical languages. Chinese texts (Han scripts) were much different than the the spoken language, it's like a bytecode or some serialization format. There are cases where verbal vocabulary not "serializable" in Chinese characters (not orthographically at least)

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

#60

Post by csa »

russellbeattie wrote: post #44

The honest truth is that despite trying to read up on the topic, I still don't understand how Chinese is used on a daily basis. Even the most basic "alphabetization" is based on strokes of the characters I guess? And Pinyin is mind boggling to me. I guess I'll just have to actually sit down and learn some Chinese - like a course or something so I can work it out in my head. I have this sense that touch screens and th…

> Even the most basic "alphabetization" is based on strokes of the characters I guess?

That is one way characters and words can be ordered. Another is pinyin order.

> And Pinyin is mind boggling to me.

It’s the English alphabet without “v”. What’s mind boggling about it?

> I have this sense that touch screens and the smartphones must have had an outsized impact on societies with character based languages, with prediction making it orders of magnitude easier to enter text.

IMHO, computers had the outsized impact on printing and publishing in general.

I think smartphones and touch screens just built on that.

The prediction stuff is nice, but I don’t think it had as big of an impact as computer text in general. Big for sure, but not as big. I guess it depends on the evaluation heuristics.

> Chinese means you can fit a ton more info onto a screen

Yes. Chinese can be very information dense. It feels faster to read than English, and I’m a fast reader in English.

> AND there's no need for icons, as the entire written language is basically icons.

Hmmm… I’m not sure I would frame it that way. The characters are logograms. Icons are ideograms. They are different.

> That means learning how to use apps and menus is marginally easier.

Definitely not the case at all in my experience.

Post reply on HN