W3C home > Mailing lists > Public > public-i18n-core@w3.org > April to June 2011

RE: [css3-lists] Remaining feedback on the module

From: Koji Ishii <kojiishi@gluesoft.co.jp>
Date: Wed, 20 Apr 2011 15:57:56 -0400
To: Tab Atkins Jr. <jackalmage@gmail.com>
CC: www-style list <www-style@w3.org>, WWW International <www-international@w3.org>, "public-i18n-core@w3.org" <public-i18n-core@w3.org>
Message-ID: <A592E245B36A8949BDB0A302B375FB4E0AC2874D35@MAILR001.mail.lan>
> > ** 4.3.4. The Japanese "spoken-out" counter styles
> > http://dev.w3.org/csswg/css3-lists/#japanese-counter-styles

> >
> > * Digit 0 is U+3007, not U+96F6 for both "japanese-informal" and "japanese-financial".
> I've had specific feedback from another Japanese speaker (from my
> team) that U+96F6 is more appropriate for both of those.  I'll just
> mark it as an issue for now.

I understand that. I had a discussion about this in Japanese W3C ML in January[1], and people split for zero. Zero has two usage: 1) to represent "0", and 2) to represent a decimal (like "0.5".) Other cases do not exist because we remove all "0" in rule 5.

There are two reasons I chose U+3007 for digit 0:
* Both U+96F6 and U+3007 are correct for "0", but using U+96F6 for a decimal is wrong.
* OOXML/ODF use U+3007.
Can you talk this to him and see what he thinks?

> > * Are we dropping "cjk-ideographic" which was in CSS 2.0?
> It would be ideal if the value
> can map to "*-informal" depending on language tag.
> Yes; the only purpose of that type seemed to be to host the cjk
> algorithm.  Since the three languages all actually have slightly
> *different* algorithms, it didn't make sense to keep it.

Thanks, I'll take this to Japanese ML, as both Mozilla and WebKit have implemented this value, although I'm not sure if they're interoperable or used widely.

> > * I can't find CJK predefined counter styles in this section. Is this because you haven't
> determined how to define the "type" property for these styles? I think we have to add
> values like "chinese" or "japenese" here because their algorithms are different. I
> understand some people do not like scripts/language names appearing as value names,
> but I don't know what else options we have. Other possibilities I can think of are:
> > - "cjk-japanese", "cjk-simp-chinese", etc.: no differences?
> > - make the value name descriptive for each algorithm, like
> "cjk-remove-all-zero-and...": I think this is preferred method in CSS, but I don't know if
> we can come up with good names in this way.
> I'm not sure what you mean.  The "spoken-out" styles are explicitly
> defined in sections 4.3.3, 4.3.4, and 4.3.5.  Then there are more
> styles defined in Appendix A, for the ones that can be expressed in
> terms of the common algorithms.
> If I'm missing any that you consider important, just let me know and
> I'll be glad to add them.
> Regarding naming, I'm open to better names.  The current names are
> mostly just what was already in the spec; I've changed a few, but I'm
> definitely not attached to any of the names.

Ah, ok, so if I understand correctly, "japanese-informal" for instance is a predefined counter style that cannot be expressed by @counter-style rule, correct? I thought all predefined styles can be expressed by @counter-style rule and they're in Appendix A.

Assuming the above understanding is correct:

* Can you add suffix and negative sign for "The Japanese "spoken-out" counter styles"? It might also apply to other predefined styles as well. I thought Appendix A is the place to spec about suffix and negative signs.

* Does that mean, authors can change suffix using
    content: counter(item, japanese-informal) "suffix"; counter-increment: item; }
but cannot customize negative signs? I hope it's possible, as authors may want to use different signs as written in the previous mail. I'm not sure how we can do this without adding "japanese-informal" to the "type" property.

[1] http://lists.w3.org/Archives/Public/public-html-ig-jp/2011Jan/thread.html#msg7

Received on Wednesday, 20 April 2011 20:02:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:23:07 UTC