W3C

– DRAFT –
Chinese Layout Task Force Teleconference

05 July 2023

Attendees

Present
Bobby, Eiso, Eric, xfq, Yijun, Zhengyu
Regrets
-
Chair
xfq
Scribe
xfq

Meeting minutes

Welcome Eiso Chan

Eiso: I know many of you

Eiso: I am based in Beijing

Eiso: I work at Culture and Art Publishing House

Eiso: I'm also an expert in the IRG and the Unicode CJK & Unihan Group

Eiso: I do a lot of character encoding stuff
… I work on digitization at the publishing house
… including the web and WeChat public accounts

[Eiso introduces his work]

[css-counter-styles-3] fallback for cjk-earthly-branch and cjk-heavenly-stem

w3c/csswg-drafts#8975

[xfq introduces the background and current status]

xfq: We discussed a related issue a few years ago, this time to discuss what the fallback should be for cjk-earthly-branch and cjk-heavenly-stem

xfq: currently it's decimal, but CSSWG wants to change it to cjk-decimal

xfq: Personally I find cjk-decimal a bit weird, we don't usually use 一一 for 11

xfq: 一一 looks like a broken two-em dash

https://w3c.github.io/clreq/#id82

xfq: simp-chinese-informal and trad-chinese-informal look more natural

xfq: any comments from the group?

Eric: For the numbers, simp-chinese-informal and trad-chinese-informal are the same, right?

xfq: For numbers, yes. But the negative sign is different (负 vs 負).

Yijun: the -informal counter styles are not necessarily used in informal scenarios, it feels a bit strange to me

xfq: The name can be changed, but that's another issue

Eric: I agree that cjk-decimal is better than decimal as a fallback
… what do others think?
… But I also find cjk-decimal weird
… We usually don't use this

Yijun: I have seen this kind of usage
… 一〇一 for 101, for example

xfq: Do we think it's ok to use cjk-decimal as a fallback? Or do we need to discuss with CSSWG?

Eric: I personally think the -informal counter styles are more natural
… but still, cjk-decimal is better than decimal as a fallback

Yijun: How do Japanese folks use it?

Eric: they don't use cjk-earthly-branch and cjk-heavenly-stem

Eric: they use iroha
… All in all, our conclusion is that cjk-decimal is better than decimal, and it would be even better if the -informal counter styles could be used.

xfq: Because cjk-earthly-branch and cjk-heavenly-stem do not distinguish between Traditional and Simplified Chinese, it is not easy to correspond.

Eric: Then let's use cjk-decimal. It's a fallback mechanism anyway, although it looks strange, Chinese-speaking people can understand it.

Eric: Two digits are especially weird, three digits are better.

xfq: I will express our thoughts to the CSSWG.

Go through the pull request list

https://github.com/w3c/clreq/pulls

w3c/clreq#553

[xfq introduces the background of this issue]

Eric: There are prerequisites for using fully justified text: line length should be multiples of the character size, and the layout engine must support punctuation width adjustment as in https://w3c.github.io/clreq/#punctuation_width_adjustment

[Discuss how to adjust the line length]

Zhengyu: Traditional Chinese is better suited for full justification than Simplified Chinese, because punctuation marks used in Taiwan and Hong Kong are usually positioned in the vertical and horizontal center of the square space left for them.

xfq: If it doesn't look good, we'll hold off on this change for now.

Eric: agreed

Zhengyu: Is it possible to change only Traditional Chinese?

Eric: I think it's either nothing or both

xfq: I agree with Eric

Zhengyu: what's the implementation status of punctuation width adjustment?

xfq: no changes recently

Zhengyu: I also agree not to fully justify the text for now. Another problem is that the Traditional and Simplified Chinese fonts are different, and the rendered results like line length may be different in some cases, which will confuse people.

xfq: OK, I'll close the PR

w3c/clreq#551

All: OK to merge

Go through the issue list

https://github.com/w3c/clreq/issues

w3c/clreq#556

[Discuss how to update this note]

[Discuss the requirements of 'align with the grid' and 'align with the border of the type area']

Bobby will comment on the issue

w3c/clreq#555

Eric: I thought about this issue

Eric: jlreq talks about the maximum line length, not the best line length

Eric: the best line length depends on the font

xfq: Do you think it is necessary to mention this in the clreq document?

Eric: If we specify a specific value, then we need to verify whether this value is reasonable

[Discuss specific numbers in horizontal writing mode]

Bobby: I think it can be mentioned in a note

Yijun: If everything is typographically perfect, except that there are 100 characters on a line, that won't look good

Eric: jlreq does not mention the normal range, they only the maximum value
… I'll try to write some text and we can discuss it

Next teleconference time

August 8 (Tuesday), 19:00-20:00 (UTC+8)

Minutes manually created (not a transcript), formatted by scribe.perl version 217 (Fri Apr 7 17:23:01 2023 UTC).