W3C home > Mailing lists > Public > www-international@w3.org > January to March 2013

Re: [text-decor-3] Emphasis marks and ruby

From: MURAKAMI Shinyu <murakami@antenna.co.jp>
Date: Mon, 04 Feb 2013 02:33:41 +0900
To: fantasai <fantasai.lists@inkedblade.net>
Cc: "www-style@w3.org" <www-style@w3.org>, CJK discussion <public-i18n-cjk@w3.org>, 'WWW International' <www-international@w3.org>
Message-Id: <20130204023341.BF30.C598BCD7@antenna.co.jp>
fantasai <fantasai.lists@inkedblade.net> wrote on 2013/02/03 5:54:29
> On 02/01/2013 10:33 PM, MURAKAMI Shinyu wrote:
> >>     * Koji has found examples where the dots are instead placed as
> >>       close as possible to the base text, so immediately above base
> >>       characters without ruby, above the ruby for characters that have it.
> >
> > I think this behavior is much better than to hide ruby in many cases.
> 
> OK. So, there are several things we could do with the spec, which of these do
> you recommend?
> 
> Question 1:
> 
>    Should we remove allowing to hide ruby when emphasis marks are present?
>    a) yes
>    b) no

a) yes.

The ruby text is a part of the document content and the emphasis marks 
are a style. Dropping content text is worse than dropping styles.

> 
> Question 2:
> 
>    What should the spec say wrt emphasis mark position when ruby is present?
> 
>    a) Require placing emphasis marks as close as possible to base text
>       (Antenna House default behavior)
> 
>    b) Allow either placing emphasis marks as close as possible to base text
>         http://nadita.com/test/ruby/ruby-kenten-ahf.png
>       or placing emphasis marks at consistent level
>         http://www.w3.org/TR/2013/WD-css-text-decor-3-20130103/text-emphasis-ruby.png
> 
>    c) Don't say anything about emphasis mark position when ruby is present:
>       leave this totally undefined.

a)

There are both cases, but a) is simpler and if we leave this undefined,
we lose interoperability.

> 
> > We (Antenna House) implemented this behavior by default and also
> 
> Koji was wondering if you know what other implementations do by default?

No. (I tested InDesing and found they place emphasis marks as close as 
possible to base text, but overlap with ruby by default.)

> 
> > implemented the -ah-text-emphasis-offset property that specifies
> > minimum offset between emphasis dots and the base text.
> >
> > I put an example:
> > http://nadita.com/test/ruby/ruby-kenten.html
> >
> > The result rendered by Antenna House Formatter:
> > http://nadita.com/test/ruby/ruby-kenten-ahf.png
> 
> Interesting. Have you considered an 'auto' value for text-emphasis-offset?
> Then you could request consistent positioning without knowing ahead of time
> whether there is ruby or not.

Yes. I want the 'auto' value, we have not yet implemented.

> 
> Also, a question: is 'text-emphasis-offset' needed for other reasons,
> or is it only used for avoiding conflicts with ruby?

We implemented it for other reason. When we cannot use the 
KentenGeneric font, we have to use normal Japanese fonts for
emphasis dots. For example, MS Mincho font have SESAME DOT
U+FE45 but that is already half-sized and centered in the 1em
square, and the following settings are needed:

  -ah-text-emphasis-font-family: MS Mincho;
  -ah-text-emphasis-font-size: 1em;
  -ah-text-emphasis-offset: -0.25em;

After that, we found -ah-text-emphasis-offset can be also used for 
the emphasis marks and ruby case :-)


Regards,


Shinyu Murakami
Antenna House
Received on Sunday, 3 February 2013 17:34:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 3 February 2013 17:34:05 GMT