Hi dear admin(s) of 開放詞典,

I am writing to bring to your attention an issue with the romanization spelling used in the Cantonese section of the 開放辭典. It has come to my attention that the current system used is the Yale scheme, which can cause some distortion in the conversion of certain pronunciation elements to Jyutping.

Share some references from other research:

1. 耶魯拼音中，/a/ 喺開音節用 a 表示，閉音節用 aa。/ɐ/ 喺閉音節中用 a 表示，無法表示開音節 /ɐ/。
2. 耶魯拼音用 j ch y 嚟代表 /ts/、/tsʰ/、/j/ 三個音。因此元音 /yː/ 需要用 yu 嚟表示。
3. 耶魯方案唔區分 /ɵ/ 同埋 /œ/，兩者都用 eu 嚟表示。亦都因爲噉耶魯拼音無法表示 /ɛːu/ 呢個音。
4. 耶魯用附加符號 ˉˊ 同埋字母 h 嚟表示聲調（新版本加入咗數字標調）。

	睡 /sɵi˨/	掉 /tɛːu˨/	𠰲 /œt˨/
耶魯	seuih	無法表示	無法表示
粵拼	seoi6	deu6	oet6

To ensure consistency and accuracy, I would like to suggest that the 開放詞典 convert all the original data in Yale to standard Jyutping. In addition, it would be helpful to use Jyutping for new words and phrases to avoid any distortion in the conversion between the two spelling systems.

I hope you can consider my suggestion to improve the overall user experience of the 開放詞典, and I appreciate your efforts in providing such a valuable resource for Cantonese language learners.

Thank you for your time and consideration.

Best regards,

CHUNG

On the one hand, the syllable m1 is spelled as m¯ (m with macron) in Yale, which follows the convention in using the macron as the mark for tone 1. On the other hand, ng1 is spelled as ñg in Yale, which may be a convenient replacement for n¯ (n with macron) that is not a single character in Unicode. However, it can confuse readers not familiar with this tilde macron shorthand. Thus, I suggest we stick to the conventional notation and use n¯ instead.

Edit: I also noticed an issue with the encoding of the macrons. It is currently encoded as a separate glyph U+00AF MACRON but I think it should be U+0304 COMBINING MACRON instead. So m¯ should be m̄, and n¯g should be n̄g instead. The font used by Github shifts the combining macron to the right but in fonts that support the symbol, the macron should be right on top of the letter.

For some vowels like iu, ui, eu, pingyambiu places the tone marks on the second letter. This may be a habit from hanyu pinyin where the tone marks are placed on the main vowels. However, the convention in Yale is to put the tone mark at the first vowel letter, not on the main vowel. This is evident in the textbook Speak Cantonese which first introduced the Yale romanization in 1960. I found a digitalized copy of the book and confirmed that indeed all tone marks are placed on the first vowel. Other dictionaries like CC-Canto also observe this convention.

biu and bui seen in the vocabulary section of the book (open digital version):

cheung and cheut seen in the vocabulary section of the book (open digital version):

Proposed updates

Here's a list of 127 Yale syllables I found in the pingyambiu that can be updated. If you find this useful, I can make a PR with these changes.

biū -> bīu
biú -> bíu
buī -> būi
cheūn -> chēun
cheún -> chéun
cheùhn -> chèuhn
cheūt -> chēut
chiū -> chīu
chiú -> chíu
chiùh -> chìuh
cheūk -> chēuk
cheūng -> chēung
cheúng -> chéung
cheùhng -> chèuhng
cheúhng -> chéuhng
deūn -> dēun
deún -> déun
dēeu -> dēu
deeuh -> deuh
diū -> dīu
diú -> díu
diùh -> dìuh
deūk -> dēuk
deūng -> dēung
ēeu -> ēu
fiū -> fīu
fiú -> fíu
fuī -> fūi
fuí -> fúi
giū -> gīu
giú -> gíu
geūng -> gēung
hiū -> hīu
hiú -> híu
hm¯ -> hm
hñg -> hng
heūng -> hēung
heúng -> héung
yeún -> yéun
yiū -> yīu
yiú -> yíu
yiùh -> yìuh
yiúh -> yíuh
yeūng -> yēung
yeúng -> yéung
yeùhng -> yèuhng
yeúhng -> yéuhng
kiū -> kīu
kiú -> kíu
kiùh -> kìuh
kiúh -> kíuh
keúng -> kéung
keùhng -> kèuhng
keúhng -> kéuhng
kuí -> kúi
leūn -> lēun
leún -> léun
leùhn -> lèuhn
leút -> léut
liū -> līu
liú -> líu
liùh -> lìuh
liúh -> líuh
leūk -> lēuk
leúk -> léuk
leūng -> lēung
leúng -> léung
leùhng -> lèuhng
leúhng -> léuhng
mēeu -> mēu
meeuh -> meuh
miū -> mīu
miú -> míu
miùh -> mìuh
miúh -> míuh
muī -> mūi
muí -> múi
muìh -> mùih
muíh -> múih
niū -> nīu
niú -> níu
niúh -> níuh
neúng -> néung
neùhng -> nèuhng
pēeu -> pēu
peeuh -> peuh
piū -> pīu
piùh -> pìuh
piúh -> píuh
puí -> púi
puìh -> pùih
puíh -> púih
seūn -> sēun
seún -> séun
seùhn -> sèuhn
seūt -> sēut
siū -> sīu
siú -> síu
seūng -> sēung
seúng -> séung
seùhng -> sèuhng
seúhng -> séuhng
teúhn -> téuhn
tiū -> tīu
tiú -> tíu
tiùh -> tìuh
tiúh -> tíuh
weūk -> wēuk
wuī -> wūi
wuí -> wúi
wuìh -> wùih
wuíh -> wúih
jeūn -> jēun
jeún -> jéun
jeūt -> jēut
jēeu -> jēu
jeeuh -> jeuh
jiū -> jīu
jiú -> jíu
jeúk -> jéuk
jeūng -> jēung
jeúng -> jéung
jeúhng -> jéuhng

Dear admin(s) of 開放詞典,

I hope this message finds you well. I would like to kindly request that Jyutping be made the default displayed spelling system of the website. As the use of Cantonese Jyutping continues to grow, many online resources have standardized around Jyutping as the primary spelling system for Cantonese. Even Apple QuickType and Google Gboard now offer Jyutping as the primary spelling system for Cantonese input methods.

I have noticed that some users have experienced inconsistencies in their use of both 開放詞典 and other online resources due to the Yale scheme currently used on the website, which is not the standard Jyutping spelling system they are familiar with.

The case shows this visitor/language learner was questioning the spelling using in site was "wrong"/non-standard, and he was unable to type the word by typing in the way it was spelled as shown:

Therefore, I would like to kindly request that you consider making Jyutping the primary default displayed spelling system for the website to better meet the needs of most users.

Thank you for your efforts on the Cantonese language and for providing such a great source of data for those who want to learn Cantonese and build applications related to Cantonese learning. I look forward to hearing back from you soon.

Best regards,

CHUNG

As defined in various sources, the Yale romanization uses numbers 1-6 exclusively. Number 1, 3, and 6 are used for 上陰入, 下陰入, and 陽入. Please consider replacing tone numbers 7, 8, 9 in Yale with 1, 3, 6. Thanks for developing this awesome lookup table.

Wikipedia entry on 耶魯拼音 (Retrieve Jun 9, 2022)

Chapter on Cantonese Romanizaton in Routledge Encyclopedia of the Chinese Language

"The peculiarity as a result of using tone letter h causes some later versions to discard it altogether and replace the diacritics with tone number: /55/ (or /53/, /5/), /25/, /33/ (or /3/), /21/, /23/, and /22/ (or /2/) are numbered from 1 to 6. In this ‘simplified’ version, yuht yúh and gwóng dūng wá are given as yut6 yu5 and gwong2 dung1 wa2" (emphasis mine).

粵拼與其他粵語拼音方案對照表 on jyutping.org

耶魯拼音一般用聲調符號嚟標調，並且喺陽聲調嘅元音後邊加 h。喺唔方便用聲調符號嗰陣，亦會使用數字標調
粵語廣州話嘅陰平調有 53 同 55 兩種調值，又分別叫「上陰平」同「下陰平」。現代粵語中呢兩種調值一般爲自由變體，冇特別區分，所以粵拼、廣拼、教院方案都用調號 1 代表陰平調；而耶魯、劉錫祥就有區分呢兩種調值。對於三個入聲調，教院方案可用 7、8、9 標，亦可用 1、3、6 標。

Solution

I wrote three simple regular expressions in JavaScript style to replace the tone numbers in pingyambiu.

1. Replace ^([a-z]+)7\t with $11\t
2. Replace ^([a-z]+)8\t with $13\t
3. Replace ^([a-z]+)9\t with $16\t