Closed Bug 916647 Opened 11 years ago Closed 8 years ago

[B2G][Contacts]: enable sorting order for zh-TW

Categories

(Firefox OS Graveyard :: Gaia::Contacts, defect)

ARM
Gonk (Firefox OS)
defect
Not set
normal

Tracking

(tracking-b2g:backlog)

RESOLVED WONTFIX
tracking-b2g backlog

People

(Reporter: CocoMo, Assigned: petercpg)

References

Details

(Keywords: feature)

Hi Peter,

I am reaching out to you to help gather necessary information to help define  the behavior in Contacts to sort names both in Chinese and English (or Latin letter based names).  Please copy any team members who may have knowledge in this area so we can ensure the needs are covered. 

- is it based on number of strokes or on Zhuyin?
- what is the sequence of tones?
- sorting order of characters of the same Zhuyin and tone?
- what are the common input methods in mobile space to enter Chinese characters?
- On the right side of the Contact list, currently it shows the 26 English letters.  What is expected for a phone with lots of Chinese contacts?  
- sorting order with mixture of Chinese contacts and Latin letter based contacts? (for instance my name shown in Chinese characters vs in English letters, where can you find me in the list).
- what if the entry contains symbols (number or #?$*, etc, as in Music album, song title or artist name), what is the sequence between this and Chinese names/titles and Latin letter based names/titles?

What are the standards/rules used these days especially in the mobile space in Taiwan?  

Help us define the requirement.  This will also help us in the testing phase.

Thank you,
Peiying
Assignee: nobody → petercpg
Due to there's no zh-HK locale so users in Hong Kong often have no choice but using zh-TW. 
I'll try to consider some basic usage of HK users. 

(In reply to Peiying Mo from comment #0)
> Hi Peter,
> 
> I am reaching out to you to help gather necessary information to help define
> the behavior in Contacts to sort names both in Chinese and English (or Latin
> letter based names).  Please copy any team members who may have knowledge in
> this area so we can ensure the needs are covered. 
> 
> - is it based on number of strokes or on Zhuyin?
Depends. For iOS, it sorts by strokes ascend (1 top, N bottom). Android (HTC Sense UI and AOSP UI) converts contact's name to pinyin, then sort by the first letter in A to Z. 
Windows Phone (WP 7.8) like Windows, sorts in strokes.
Both iOS and Android provide options for users to sort by first or last name, and which comes first.

> - what is the sequence of tones?
I guess you are talking about tonal marks? https://en.wikipedia.org/wiki/Bopomofo#Tonal_marks
Like mentioned in wikipedia, we use marks to distinguish 5 tones in Mandarin. 
Zhuyin has 4 marks, (null) is used for first, ˊ for second, ˇ for third, ˋ for fourth, and ˙ for light tone.

> - sorting order of characters of the same Zhuyin and tone?
Since I don't have any mobile platform that sorts in Zhuyin, I have no answer about this. Sorting in tone 1 -> 5 would be ok to me.

The only platform sorts filename in Zhuyin I can test is Windows (it's set to strokes by default), and I can't find a rule.
Testcase: (Sorted in File type, which are all "Text documents")
Rank	Name	Zhuyin	Pinyin
1	低.txt	ㄉㄧ	di
2	底.txt	ㄉㄧˇ	di3
3	的.txt	ㄉㄧ˙	di5
4	迪.txt	ㄉㄧˊ	di2
5	第.txt	ㄉㄧˋ	di4

Please note that one word could be mapped to different readings, some might have only tones changed with Zhuyin alike, while some might have a totally different Zhuyin.
Not sure about which sound is picked to sort in Windows when there's more than one.

> - what are the common input methods in mobile space to enter Chinese characters?
Except handwriting and voice input, Zhuyin and Changjie/Quick Changjie are bundled in most mobile devices. Some users installs their accustomed IMEs such as Dayi, Array, and Boshiamy. (might not be all available in all platforms)
A marketing research said that in Taiwan market, the usage of Zhuyin is about 70%, while Changjie, Boshiamy and others share ~10% each.

My opinion is that we can support Zhuyin (taught in Taiwanese schools after 1949 so most people can use) for first and
Changjie/Quick Changjie (for users don't read Zhuyin, mostly Hong Kong users, or seniors).

> - On the right side of the Contact list, currently it shows the 26 English letters.  What is expected for a phone with lots of Chinese contacts?
The Zhuyin characters list of the first word of each entries, I believe. However there're 37 characters maybe a table, or a switch then a bar that won't disappear would be good? 
This case might be a common issue to Japanese, there're 48 Katakana and Hiragana characters. Counting strokes is a pain that I can hardly count it right :/
Working like Android is also applicable and somewhat more neutral but is less desired to me.

> - sorting order with mixture of Chinese contacts and Latin letter based contacts? (for instance my name shown in Chinese characters vs in English letters, where can you find me in the list).
Alphanumerical first, then Chinese, vice versa.

> - what if the entry contains symbols (number or #?$*, etc, as in Music album, song title or artist name), what is the sequence between this and Chinese names/titles and Latin letter based names/titles?
As above, different sequeces are ok to me. When symbols are put in the first char of entry name you can treat it like Latin, otherwise it just depends on the first Chinese word.
There're also symbols/punctuations in Chinese (and differs from what are used in China). Most are same to English while the width or position is different.
Example:
quotation marks	-> 「」
semicolon -> ;
ellipsis -> …

> - What are the standards/rules used these days especially in the mobile space in Taiwan?  
Not sure about this. any examples?

> 
> Help us define the requirement.  This will also help us in the testing phase.
> 
> Thank you,
> Peiying
Hi Peiying,

Do we have a targeted B2G version here or this is simply a research? Sort a list (in any way _other_ than the Unicode order) is non-trivial development work since JavaScript does not currently offer any locale sorting functions natively.
Flags: needinfo?(pmo)
Does having the js i18n api/ICU help? That'd give us a front end lever to push bug 866301. Docs for that are at http://www.ecma-international.org/ecma-402/1.0/#sec-10

Please bubble up your management chain to get information regarding markets and milestones.
@Tim, we will have a B2G version soon, but sorting issue needs to be resolved.

@Peter, thanks for providing additional info that clarifies and adds more to the complexity of the issue.  Regarding the question - 
> - What are the standards/rules used these days especially in the mobile space in Taiwan?  
Not sure about this. any examples?

I am more familiar with the printed docs and how the index works for Traditional Chinese.  I was not clear which rules to follow or are more widely implemented in mobile. That's why I raised the question.  I guess, do we have to make all options available or just one or two in the initial introduction of the phone.  

On the tones, the one  的.txt	ㄉㄧ˙	di5 (or the 5th tone) is usually used at the end of a sentence, and the chance of this as the first name is probably unlikely, not to mention 的 has more than one pronunciations and they mean different things.
Flags: needinfo?(pmo)
I expect that we'll need the js globalization API collation stuff to really fix this. Marking this as blocked by bug 866301.
Depends on: 866301
blocking-b2g: --- → backlog
blocking-b2g: backlog → ---
This is no longer relevant.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.