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

Japanese explained to programmers

lajili.com

Post reply on HN
Print view View original

401–410 of 416 posts

Re: Japanese explained to programmers

#401

Post by Ferret7446 »

philsnow wrote: post #196

The author has discovered (linguistic) morphology. The `setRole` function he describes compiles to various morphemes in Japanese like は, を, etc. https://en.wikipedia.org/wiki/Morphology_(linguistics) We have this in English too except remarkably little of it actually comes out. Pluralization and possession being marked with -(e)s and -'s suffixes, and "subject-verb agreement" (one wouldn't say "the birds eat s the fo…

I would hazard a guess that Japanese is the second natural language the author has learned. When you learn your second language, you see the aspects of a language that you never explicitly learned with your first language, and additionally are able to connect that knowledge back to your first language.

Re: Japanese explained to programmers

#402

Post by barking_biscuit »

wst_ wrote: post #366
barking_biscuit wrote: post #267

Earlier quoted context omitted.

>The "many pronunciations for one kanji" thing isn't a problem in practice, because 1) most kanji have _1_ on'yomi (Chinese) and _1_ kun'yomi (native) pronunciation, with it being obvious which one to choose from context, especially because when it's kun'yomi, the kanji are suffixed with hiragana to denote its correct pronunciation. There are however problematic kanji, such as the one you note, because there are seve…

Isn't it common in English, too? Just to mention "vague", "awe", "sigh". There are many others. I see those words and pronunciation is far from obvious to me.

It is, but in Japanese it's on a whole other level. One key thing a lot of people don't really account for is the fact that native speakers of a language are already fluent in their language before they learn to read, and in many cases (certainly not all) have the added benefit of having heard the word before and the task is just to create a mapping between the sound and it's visual representation (for words already in their vocabulary). When you're a non-native speaker, you don't have this advantage and the difference it makes can be quite large. So, it's natural for anyone non-native speaker to struggle with this. The difference is in frequency, severity and duration. You can get a pretty good hint at how much insanely harder it is for non-native Japanese speakers learning to read Japanese by comparing how much native speakers both kids and adults struggle to read unfamiliar words. I can't really ever recall struggling to read unfamiliar words much in English, and I compare that with my wife who is Japanese and how many times we have played a guessing game at how to read an unfamiliar Japanese word in a book she was reading. It took me a long time to realise that even for Japanese people there's just this level of insurmountableness to it that is part and parcel of how the orthographic system is structured. Going back to how kids learn to read unfamiliar words, books in Japanese have furigana in them which are a phonetic guide on how to read words that the author assumes the intended audience has a high chance f not being able to read. Even general purpose books aimed at adults have quite a lot of this. The younger the intended audience, the more furigana tends to be present as it's simply not possible to read without it when you're much younger. What you often see, though not always, is for a difficult word to have furigana the first time it appears and then subsequent appearances of the word don't have it. This is one way you learn to read more over time. For certain words, no matter how many times it appears, if it's obscure enough it may always appear with furigana.

The existence of furigana alone is a huge admission that Japanese a somewhat unreadable language that has somehow or other been made to work through bolting it on. So, you can imagine that if an orthographic system comes with permanent training wheels for native speakers, then non-native speakers are going to have a pretty brutal time with it.

Re: Japanese explained to programmers

#403

Post by Tor3 »

kibwen wrote: post #379
Tor3 wrote: post #358

Earlier quoted context omitted.

I can read only a limited number of kanji (or rather, words written with kanji), but even that makes reading much easier. Books for small children are written in Hiragana, with spaces, and in theory that should be as easy to read as English.. after all, it's easy to learn Hiragana, anyone can do that in a short time, and with practice you can learn to read it fast. But it's still tricky to read all-Hiragana sentences…

It's interesting to consider that Japanese's relatively low number of discrete phonemes might result in a relatively high number of homophones, but if a preponderance of homophones is the problem, wouldn't listening to spoken Japanese be just as hard as reading hiragana?

On the contrary, listening to Japanese feels easier than listening to a lot of other languages, even for speech which you don't actually understand. The sounds are simple, there are a lot of vowel sounds. Straight forward vowels, not diphthongs, and in that respect it's a lot like listening to (slow) Italian. There are a few run-on consonants in speech, but not many. Of course one runs into homophones in speech too, but there you also sometimes (but not always) have pitch to help you (persimmon and oyster are both written かき(kaki) in Hiragana, but the pitch is different, same for chopsticks/bridge), and there's context - if you understand enough of the language.

As an example, my wife owns a Japanese CD which includes a couple of songs originally from my own country, but translated to Japanese. Listening to that is very easy, I hear every word.. it's much much harder to actually get the words if I listen to the original in my own language! And these are simple songs.

EditAdd: What I said above is correct in the sense that it's easy to listen to Japanese, as in actually hearing the words, but yes there are homophones which sometimes make me, a low-intermediate speaker, not understand what I hear. I hear "another" similar-sounding word, unless there's pitch to clearly identify it, and then what I hear doesn't make sense unless it's a recording and I can re-listen.

Re: Japanese explained to programmers

#404

Post by iopq »

g9yuayon wrote: post #395
iopq wrote: post #354

Earlier quoted context omitted.

The Spanish sentence is exactly like the English sentence "Programming is pleasing to me", it's the same grammar and it exists

The structure is different, right? "Me gusta algo" does mean something is pleasing to me, but it's not the way how one would think in Chinese or English.

But the grammar is the same as "something pleases me", no difference at all

But there's other verbs, just like in English that you can use like this:

Disfruto de pasar tiempo con mi familia.

I enjoy spending time with my family

Re: Japanese explained to programmers

#405

Post by antiterra »

midoridensha wrote: post #341
antiterra wrote: post #191

I recognize that one can’t expect English speakers to consistently pluralize foreign words in ways that match the origin language, and that it is not objectively incorrect when people just add an s. However, in my opinion, an article purporting to inform about Japanese in a programmatic way really should avoid using “hiraganas,” “kanjis” and “katakanas” as pluralization of thise terms.

Those are just plain wrong. It's like talking about "English alphabets". There's only one. What they mean is "hiragana characters", etc.

It looks like it’s been edited and is far more agreeable to the idea that ‘hiragana’ et al are already plural.

Re: Japanese explained to programmers

#406

Post by bamboozled »

ezconnect wrote: post #398
ickoonite wrote: post #397

Earlier quoted context omitted.

The stereotype of Japan as technologically advanced persists in the West despite abundant evidence to the contrary. Japan still uses floppy disks and fax machines. They prefer to pay in cash and will go to the bank to get crisp, fresh notes - to give old notes would be disrespectful. Banking in Japan - indeed, experience with any kind of bureaucracy at all there - makes the US DMV, or indeed any Western bureaucracy,…

I live in Japan, they do love cash and I do too (less government control and a safety haven) but they are not way behind on tech compared to the west.

they are not way behind on tech

Can you name one, decent, enjoyable to use online Japanese store, or service? Maybe Line works ok? I can't think of anything else.

Ever tried to buy a Shinkansen ticket online ?

Re: Japanese explained to programmers

#407

Post by midoridensha »

antiterra wrote: post #405
midoridensha wrote: post #341

Earlier quoted context omitted.

Those are just plain wrong. It's like talking about "English alphabets". There's only one. What they mean is "hiragana characters", etc.

It looks like it’s been edited and is far more agreeable to the idea that ‘hiragana’ et al are already plural.

It's not plural at all; it doesn't even make sense. This is like debating whether the word "happiness" is singular or plural.

There's no such thing as "a hiragana".

Re: Japanese explained to programmers

#408

Post by iopq »

rjh29 wrote: post #389
iopq wrote: post #351

Earlier quoted context omitted.

This is just not true let's take something very basic, the kanji 人 has two on'yomi pronunciations, jin and nin and there's no way to tell which one to use without knowing ALL of the words it will be used in 日本人 - nihonjin, so far so good 三人 - sannin, okay, I guess there's two pronunciations 二人 - futari, huh? 一人 - hitori, wat so please tell me how I am supposed to know 人生 (jinsei) from 人気 (ninki) from 恋人 (koibito) fro…

You basically have to memorise them, like with any other language (e.g. the rules and exceptions of English spelling and pronunciation). Usually it's the most common words that have weird exceptions, because we use them enough to remember them. All the words you mention are beginner words. (There's also 人気=ひとけ, just for fun). As far as jin and nin goes... (source: 漢語からみえる世界と世間) jin is more about adjectives/nouns - de…

Based on your rules, you would not be able to predict 人気 ninki, 人類 jinrui

Why is popularity "plain" and humanity is "identity"?

You just have to memorize ALL of the compounds, there's no other way

Re: Japanese explained to programmers

#409

Post by ezconnect »

bamboozled wrote: post #406
ezconnect wrote: post #398

Earlier quoted context omitted.

I live in Japan, they do love cash and I do too (less government control and a safety haven) but they are not way behind on tech compared to the west.

they are not way behind on tech Can you name one, decent, enjoyable to use online Japanese store, or service? Maybe Line works ok? I can't think of anything else. Ever tried to buy a Shinkansen ticket online ?

Train tickets are the easiest thing to buy in Japan phone app, vending machine you name it. The train attendant even sells you ticket if you get on board without paying.

Re: Japanese explained to programmers

#410

Post by antiterra »

midoridensha wrote: post #407
antiterra wrote: post #405

Earlier quoted context omitted.

It looks like it’s been edited and is far more agreeable to the idea that ‘hiragana’ et al are already plural.

It's not plural at all; it doesn't even make sense. This is like debating whether the word "happiness" is singular or plural. There's no such thing as "a hiragana".

In regular usage there definitely is, it’s easy to find something like this on a Japanese website: “どの漢字がどのひらがなになったの？”

That means “Which kanji became which hiragana?” If there’s no such thing as ‘a hiragana’ that wouldn’t make sense.

As for kanji, “この漢字” also shows up all the time when referring to a specific character. References to specific amount of kanji are everywhere too, eg the “100 Kanji” here. https://www.kinokuniya.co.jp/f/dsg-08-EK-1085796

Post reply on HN