W3C home > Mailing lists > Public > www-style@w3.org > December 2010

Re: Need to clarify the effects of bidi paragraph breaks

From: Alan Gresley <alan@css-class.com>
Date: Thu, 16 Dec 2010 21:57:30 +1100
Message-ID: <4D09F09A.8040505@css-class.com>
To: Ambrose LI <ambrose.li@gmail.com>
CC: "Aharon (Vladimir) Lanin" <aharon@google.com>, W3C style mailing list <www-style@w3.org>, fantasai <fantasai.lists@inkedblade.net>, "public-i18n-bidi@w3.org" <public-i18n-bidi@w3.org>
On 16/12/2010 4:55 PM, Ambrose LI wrote:
> 2010/12/16 Alan Gresley<alan@css-class.com>:
>> That believe this depends on the value of unicode-bidi. I am somewhat
>> confused myself since the default behavior in an offline test,
>>
>> <!DOCTYPE html>
>> <div dir=ltr>
>> <span dir=rtl>
>> TO BE<br>
>>   OR NOT TO BE?
>> </span>
>> <div>-- hamlet, in rtl translation.</div>
>> </div>
>>
>>   in FF 3.6.13 renders as embed where the initial value for unicode-bidi is
>> normal.
>>
>>
>>    unicode-bidi: embed, isolate and plaintext produces this.
>>
>>   ?OR NOT TO BE
>>
>>
>>    unicode-bidi: normal produces this.
>>
>>   OR NOT TO BE?
>
> Note that Aharon had stated that
>
> TO BE
> OR NOT TO BE?
>
> actually represent actual RTL characters. So if you are going to do an
> actual test on a browser you need to replace this with actual RTL
> characters, such as
>
> <div dir=ltr>
> <span dir=rtl>
> <br>
>   או לא להיות?
> </span>
> <div>-- hamlet, in rtl translation.</div>
> </div>
>
> Otherwise the tests will not give any meaningful results.


Ah ha, now that's a clue. This also stresses my point regarding have 
words of various script that can be meaningful for someone who doesn't 
understand them. I presume this,

    להיות

is Hebrew. A Side note. As I pasted it above, the end of line is on the 
left and the start (home on keyboard) is on the right. This is new for me.

Now what does this word mean? What I would like to see is something that 
is recognizable to the eye but has meaning. I know this 漢字 which is 
written in Kanji and I believe is the characters for the word Kanji. My 
Mongolian script (done with Unicode entities &#x1828;) ᠨᠶᠪᠧᠺᠴᡗ is a run 
of Mongolian letters. I wouldn't know if it spells anything.

For me to sometimes even test anything bidi. I must copy some random 
word(s) in some script. Then to preserve the Unicode, I must create the 
test in Notepad. My usual text editor doesn't allow Unicode.

I have done various test cases with bidi but it's like shooting in the 
dark. I can equate this with an author that takes on a project with 
bi-directionally and then attempting to use the current spec and having 
to grasp many things to not lose a job.

Now testing what you have. I have learned how to copy back to front for 
the first time ever. This is what I see but with the comma on the left.

להיות
או לא להיות?


Now since I couldn't manipulate the script to see what I see render, how 
can I be sure that an error has not been introduced by copying the code 
you provided and pasting into Notepad. So at this point I can not answer 
any question of where the ? should be after break.

There must be some way that we can communicate and all understand. I 
know Unicode for  less-than and greater-than but I do not know atm what 
I need to use that allows the start of the test flow on the right.



-- 
Alan http://css-class.com/

Armies Cannot Stop An Idea Whose Time Has Come. - Victor Hugo
Received on Thursday, 16 December 2010 10:58:07 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:35 GMT