W3C home > Mailing lists > Public > public-i18n-cjk@w3.org > January to March 2012

Re: Memo from ruby disucssion with Roland

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Thu, 23 Feb 2012 04:51:55 +0100
To: Roland Steiner <rolandsteiner@google.com>
Cc: Koji Ishii <kojiishi@gluesoft.co.jp>, fantasai <fantasai.lists@inkedblade.net>, "public-i18n-cjk@w3.org" <public-i18n-cjk@w3.org>
Message-ID: <20120223045155593755.51bc32bd@xn--mlform-iua.no>
Roland Steiner, Thu, 23 Feb 2012 11:55:15 +0900:
> With "current standard" I referred to the HTML5 spec,

OK. Then we agree: HTML5 has a column-major model.

> which essentially encodes IE's implementation, and which in turn is 
> followed by WebKit's implementation:
> 
>     <ruby>base1<rt>text1</rt>base2<rt>text2</rt>base3<rt>text3</rt></ruby>
> 
> i.e., column-major, without <rb>, <rbc>, <rtc>. (<rp> is supported, 
> but omitted for simplicity.)

Actually, Webkit and IE has several differences.

> I fail to see how this can be construed as having "some support for 
> both models".

Do a comparison of how IE and Webkit handles row-major, and you will 
see what I mean:

* IE handles row-major [yes it does]
* Since IE9, <rtc> can be used
  If you use HTML5 shiv, you can use it in IE6-8 too
* IE — at least until IE9 - does not auto-close any elements 
  when it sees <rt>. [Well, there is one element that makes it
  auto close, and that is if you place a <ruby> inside a <ruby>
  - that's like trying to place a <p> inside a <p>.]

I demo all this - except the auto-closing of <ruby> inside <ruby> - 
here:

http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1348


Leif Halvard Silli

> On Thu, Feb 23, 2012 at 11:42, Leif Halvard Silli  wrote:
>> Roland Steiner, Thu, 23 Feb 2012 11:02:18 +0900:
>>> On Thu, Feb 23, 2012 at 07:24, Leif Halvard Silli wrote:
>>>> 1. Are there anyone - apart from Ian - with a stake in this, that argue
>>>> that it should be column-major?
>>>
>>> The current standard and implementations are column-major.
>> 
>> You mix it up, I think. If we take specs firsts, then according the
>> letter from Koji that started this thread,[1] then [ignoring <rtc> and
>> <rbc>] this is the XHTML Ruby module' model — 'row-major':
>> 
>>   <ruby><rb/><rb/><rb/>
>>         <rt/><rt/><rt/><ruby>
>> 
>> While this is the model that Ian placed in HTML5  — 'column-major':
>> 
>>   <ruby><rb/><rt/><rb/>
>>         <rt/><rb/><rt/></ruby>
>> 
>> If we look at implementations, as long as we with 'support' have visual
>> display in mind, then IE and Webkit appears to have some support for
>> both models. [But if we consider what they present to find-in-page,
>> screen readers or present as fallback with CSS disabled, then they only
>> support row-major.]
>> 
>> Koji's description of 'row-major': '"row-major" approach; split first
>> by rows and then by columns.'
>> 
>>>> 2. Do we agree that column-major - what is in HTML5 now - should be
>>>> non-conforming?
>>>
>>> I don't think that's an option as it would break existing pages. A
>>> solution should have graceful fallback to both the current standard
>>> as well as to no <ruby> support.
>> 
>> I believe, when you said 'column-major', your really meant 'row-major',
>> not? And if so, then we can conclude, that so far, everyone in this
>> group is in favor of row-major.
>> 
>> The question I am still uncertain of, though, is whether anyone thing
>> that HTML5's column-major needs to remain conforming. My opinion about
>> that is negative - it need not and should not.
>> 
>> [1]
>> 
http://www.w3.org/mid/A592E245B36A8949BDB0A302B375FB4E0D334EFCD6@MAILR001.mail.lan

>> --
>> Leif Halvard Silli
> 
Received on Thursday, 23 February 2012 03:52:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 23 February 2012 03:52:30 GMT