Korean-specific CSS issues to be discussed

Hello Korean friends,

My name is Kang-Hao (Kenny) Lu and I am a Chinese speaker working for 
W3C on, partially, internationalization on Web. It's very proud to see 
you having fruitful discussion about cutting edge HTML5 features and 
news (like the one about h.264 :) ). As CSS is essential to the Web but 
it is still far from complete in terms of its multilingual support, I 
would like to invite you to review CSS specs that have Korean-specific 
features. Notably,

* CSS 3 Text[1] - a CSS module for multilingual text support, currently 
under heavy revision.
* CSS 3 List[2] - a CSS module for lists. It has not been updated for 
many years but CSS WG just appointed a new editor to work on it
[1] http://dev.w3.org/csswg/css3-text/
[2] http://dev.w3.org/csswg/css3-lists/

To review the specs, basically you can just do a text search on the 
keywords "Korean", "Hangul" to find the relevant parts. Please don't be 
surprised when you find unexpected description of Korean (I was 
surprised the first time I read the parts about Chinese :) ) because the 
editors wouldn't be able to know every detail about every language, and 
that's why we need your help. There are actually several specific 
questions in these specs, but of course your discussions are not limited 
to these:

1.
For vertical writing in Korean, on which side do you put an underline 
on. The current draft says it should be put on the right (see 
'text-underline-postion: auto' [3]), but my experiment[4][5][6] with IE 
shows something different. Which is correct? A real world picture, even 
from an old book, might tell a lot.

[3] http://dev.w3.org/csswg/css3-text/#text-underline-position
[4] http://www.w3.org/People/kennyluck/Test/underline-pos
[5] http://www.w3.org/People/kennyluck/Test/underline-pos.ie7
(the forces the page to be displayed in IE7 legacy mode and you can see 
the underline for Japanese is on the right but it's on the left for Korean)
[6] http://www.w3.org/People/kennyluck/Test/underline-pos.ie8

2.
For CSS 3 List, several changes have been suggested by a researcher from 
Microsoft[7]. You are encouraged to review the document as a whole. But 
the editor has a particular concern about 'hangul-legal'. Here's our 
conversation

[[
Tabatkins: kennyluck: Thanks a lot for the pointer back to that document 
about cjk fixes!  I had forgetten about it.  Question for you guys when 
you review - is it intentional that hangul-legal doesn't have it's own 
version of the second digit marker?  Right now only the second digit 
marker, and the digits themselves, are falling back to cjk-ideographic.  
That feels like a possible error.
kennyluck: TabAtkins: I am not a Korean, but I my guess is that 
Hangul-legal uses a tweak version of the cjk-ideographic algorithm.
kennyluck: 3. For each group, ignoring digits that have the value zero, 
append the second
kennyluck: digit marker to the second digit, the third digit marker to 
the third digit, and
kennyluck: the fourth digit marker to the fourth digit. These markers 
are defined in the
kennyluck: tables for the specific numbering systems. The first digit 
has no marker.
kennyluck: Perhaps "append the second digit marker to the second digit" 
does not apply to Hangul-legal, since it does not have it's version of 
the second digit marker.
kennyluck: so for example, if 40032 is to converted to Hangul-legal.
kennyluck: the "32" part became 3 2 and then 서른 둘
kennyluck: instead of 3 + the second digit marker + 2
kennyluck: 서른 and 둘 are from the table with the caption "For values 
between 1 and 99, appropriate digits are picked from the following list 
(at most one per column) and written in descending order by value. "
tabatkins: kennyluck: If that's true (about hangul-legal not having a 
second digit marker) I'd need that clarified.  Right now my plan is to 
just treat it as all the other cjk systems, with a fallback to the 
specified additive system for numbers below 100.
]]

You might also want to think about how useful this particular list type is.

[7] http://lists.w3.org/Archives/Public/www-style/2010Feb/0153  (the 
attachement)

3.
"word-break:keep-all" seems to be used quite often in Korean, to 
disallow breaks in a series of Hangul not separated by white spaces. 
What should happen when "word-break:keep-all" is applied with 
"word-wrap: normal"[8]? (word-wrap is a property that triggers 
"emergency wrapping" and normal is its default value)

For example, if you apply "word-break:keep-all" to, say, "안녕하세요" 
and then shrink the window until the window is smaller than "안녕하세 
요". Should "안녕하세요" be split? I haven't tried it myself but you are 
encouraged to do some experiments.

Notice that the current draft says yes but previous version[9] said no.

[8] http://dev.w3.org/csswg/css3-text/#word-wrap
[9] 
http://dev.w3.org/cvsweb/~checkout~/csswg/css3-text/Overview.html?rev=1.17&content-type=text/html;%20charset=iso-8859-1#word-wrap0


You are encouraged to discuss these in Korean. After you have some 
conclusions about these issues, you are encourage to send feedback to 
the mailng list www-style [10] and Cc public-i18n-cjk [11]  list.

[10] http://lists.w3.org/Archives/Public/www-style/
[11] http://lists.w3.org/Archives/Public/public-i18n-cjk

Sorry for not being able to write this email in Korean. Translation of 
this mail will be much appreciated!

Cheers,
Kenny
W3C, Internationalization Working Group Member

Received on Wednesday, 19 January 2011 14:58:59 UTC