W3C home > Mailing lists > Public > www-style@w3.org > January 2007

Re: [CSS3 Text] word-break

From: fantasai <fantasai.lists@inkedblade.net>
Date: Wed, 10 Jan 2007 08:18:53 +1300
Message-ID: <45A3EA9D.20009@inkedblade.net>
To: MURAKAMI Shinyu <murakami@antenna.co.jp>
CC: www-style@w3.org

MURAKAMI Shinyu wrote:
> In the current CSS3 Text draft spec:
> http://www.w3.org/TR/css3-text/#word-break
> 
> Here "normal" means "strict" (disallow line-breaking before small kana 
> etc.). I think this is not very good because:
> 
> 1. "loose" (allow line-breaking before small kana etc.) line-breaking 
>    restrictions are much more often used than "strict" in current 
>    Japanese typography.
>
> 2. Some CSS implementations (for example, Internet Explorer) already 
>    implemented the "loose" line-breaking as default behavior. If the 
>    strict line-breaking becomes the default in the CSS standard, such 
>    existing implementations will be no longer standard compliant.

Some other implementations (for example, Mozilla, Opera, and Safari) have
implemented "strict" as the default behavior. I know that Kaz Otomo, who
dealt with these issues at Opera, felt that using loose line breaking by
default would be poor.

> My proposal is to separate "strict" from "normal". The new definition is 
> the following.
> 
> Name: word-break
> Value: normal | strict | keep-all | loose | break-all |  break-strict 
> Initial: normal 
> 
> normal
>     The UA determines what set of line-breaking restrictions. It may be 
>     equivalent to either of 'strict' or 'loose'.

The problem with this is that it's not interoperable. (We want CSS
implementations to be consistent, which is why 'normal' means one
thing only, not a choice of two things.)

I don't think we'll make these suggested changes, but I'll keep them in
mind if this topic comes up again.


While we're on this topic, if you could specify what breaks are disallowed
in strict that are allowed in loose, that would be very helpful information
to put in the spec. So far I have

   * breaks before small kana
   * breaks before iteration marks
   * breaks before katakana lengthening mark

I know that various combinations are possible, so I'm leaving this as a
"recommendation", but I would like to have a reasonable baseline behavior
described in the spec.

> BTW I think that the "break-strict" might be unnecessary. Such line
> breaking restrictions are very unusual in the real Japanese typography.

That was what I expected; it is there for completeness I suppose. Richard
Ishida complained when I tried to remove it. :)


Thank you for your comments~
~fantasai
Received on Tuesday, 9 January 2007 19:19:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:49 GMT