[ファブリス]

The UNIHAN database has different "reading" fields, I'm a little confused as to what is what.

For example in the character 语 ( link to mdbg.net ):

U+8BED kCantonese jyu5
U+8BED kDefinition language, words; saying, expression
U+8BED kHanyuPinlu yu3(261)
U+8BED kMandarin YU3 YU4

Does Mandarin mean Pinyin (YU3 YU4 => yǔ, yù) ?

Is Cantonese (jyu5) using the same Pinyin "romanized" system?

In this data above, is "yǔ, yù" the reading that you want to see on Chinese flash cards?

Would you use the mandarin reading on simplified chinese hanzi flash cards, and the cantonese reading for traditional hanzi flash cards, or does it not matter?

Cantonese does not have a standard romanization. Common systems are Yale, Guangdong government, and Jyutping. Yale is by far the most used system in English dictionaries and is what jyu5 refers to.

http://en.wikipedia.org/wiki/Yale_Romani...#Cantonese

ファブリス Wrote:Does Mandarin mean Pinyin (YU3 YU4 => yǔ, yù) ?

Yes

ファブリス Wrote:Is Cantonese (jyu5) using the same Pinyin "romanized" system?

Nope

ファブリス Wrote:In this data above, is "yǔ, yù" the reading that you want to see on Chinese flash cards?

I prefer that style but tone numbers would be fine too.

ファブリス Wrote:Would you use the mandarin reading on simplified chinese hanzi flash cards, and the cantonese reading for traditional hanzi flash cards, or does it not matter?

This does matter. Traditional hanzi does not equal Cantonese. You are forgetting Taiwan. If you want to, you can just put in both. But don't separate them like that because that would really piss off the people wanting to learn Taiwanese Mandarin.

If you add Cantonese pronouncation, add it to both simplified and traditional characters since most Cantonese speakers use simplified hanzi.

[ファブリス]

Thankies!

I see, UNIHAN's kCantonese field refers to the jyutping romanization; while "kMandarin" refers to the Mandarin pronunciation in pinyin.

MDBG.net character dictionary shows both Yale and Jyutping. example. I would assume they get both from UNIHAN but I don't find Yale in the UNIHAN Readings file.

Is it possible that mdbg.net derives Yale from jyutping? For example ceng1, cing1 (jyutping) => cheng1, ching1 (yale); or loeng5 (jyutping) => leung5 (yale) ?

ファブリス Wrote:Would you use the mandarin reading on simplified chinese hanzi flash cards, and the cantonese reading for traditional hanzi flash cards, or does it not matter?

I assume this is related to the RTH site you are working on. You would have to have people set as a preference which language(s?) they are interested in. There are no simple mappings in either direction. Any Chinese dialect--Mandarin, Cantonese, Classical Chinese, and others--can be written in either traditional or simplified (although it's uncommon for dialects outside of those three to be written down at all), and the precedent for doing so depends more on locale than the language itself.

ファブリス Wrote:Is it possible that mdbg.net derives Yale from jyutping? For example ceng1, cing1 (jyutping) => cheng1, ching1 (yale); or loeng5 (jyutping) => leung5 (yale) ?

Nope, http://en.wikipedia.org/wiki/Jyutping#Co...manization , and the fact that the Jyutping romanization according to wiki was created decades after Yale.

Hey Fabrice, is it possible to use the free audio databases to put sound on the flashcards on the new website?

Or then would you have to do both mandarin and cantonese so you don't show favoritism? (Mandarin should be the favorite )

[ファブリス]

mafried: I am talking about the pronounciation of the characters, as displayed on the flash cards. I thought one type of romanization may have been linked to simplified, and another to traditional and I understand that's not necessarily the case.

I was not talking about the Simplifed/Traditional choice. The first live version will be only traditional, so I am not trying to figure out yet how to handle the switch between traditional/simplified. I confirm simplified will require another database column, and a completely separate set of unicode code points (some of which may be the same as RTH, others not).

What I found in UNIHAN is "variants". That may actually allow me to show variants of one character, whether the current mode is traditional or simplified. UNIHAN defines stylistic variants (z variants: 說 and 説) and shape variants (y variants: 貓 and 猫).

How usable is pinyin, compared to other romanization methods? Could I simply use pinyin by default for the first release? If I have to pick one for the time being, which would make most sense for traditional characters?

Ryuujin27: can do, hopefully as a HTML5 audio element. This is the list I gathered after reviewing all topics here, do you have other sources? Which one is best? (EDIT: checked licenses, and none are workable, see red text).

- http://swac-collections.org/download.php (words, not character pronunciations)
- http://www.chinese-lessons.com/download.htm (NonCommercial)
- http://njstar.com/mandarin_sounds.zip (link broken, unknown license)
- http://chinesepod.com/resources/pronunciation (not a public download)

I will not add audio in the first testable version, but I will take notes for when I get around to it. I want to get a traditional hanzi version out the door asap, with no new Hanzi-specific features, except for adding one type of romanization on the flash card.

ファブリス Wrote:How usable is pinyin, compared to other romanization methods? Could I simply use pinyin by default for the first release? If I have to pick one for the time being, which would make most sense for traditional characters?

If you had to pick one pronunciation, use pinyin. It is the official romanization of Mandarin Chinese.

You can leave Cantonese out of the first release because the vast majority of people interested in this website are going to be studying Mandarin Chinese, not Cantonese. Make the most people happy.

[ファブリス]

Ty!

Let's continue the conversation here: Reviewing the Hanzi: Development Log.

ファブリス Wrote:mafried: I am talking about the pronounciation of the characters, as displayed on the flash cards. I thought one type of romanization may have been linked to simplified, and another to traditional and I understand that's not necessarily the case.

Oh, my bad :\ sorry.

I'll add my other feedback to the other thread.

[ファブリス]

I'm still a bit confused with the Yale and Jyutping distinction. Are these both still widely in use for Cantonese Chinese?

Hi ファブリス,
I saw this thread while googling for something else.
Both Yale and Jyutping romanisation systems are still in used, usually in dictionaries. As for place names in Hong Kong, another Cantonese system is ultilised and it doesn't quite accurately describe the change of Cantonese phonology.
Jyutping is more common for native Cantonese speakers while Yale is more for foreigner Cantonese learners.

There are other romanisations for Cantonese but Yale and Jyutping are by far the most common nowadays in dictionaries and learning books.

Here's the comparison tables among all the romanisations:
Initial (beginning consonant):
http://arts.cuhk.edu.hk/Lexis/lexi-can/initials.php
Final (vowel or vowel+ending consonant)
http://arts.cuhk.edu.hk/Lexis/lexi-can/finals.php
Tone:
http://arts.cuhk.edu.hk/Lexis/lexi-can/tones.php

*LSHK is another name for Jyutping

Jyutping and Yale are quite similar except for a few consonants and vowels.
Jyutipng usually tries to mimic Mandarin pinyin.
for example, "ch" in Yale is "z" in Jyutping, which is same as Mandarin pinyin "z";

If you have more question, you can ask me in my FB page.
http://www.facebook.com/hkcantonese#!/hkcantonese

J40