W3C home > Mailing lists > Public > www-style@w3.org > October 2003

CSS21 9.10 Text direction

From: Tex Texin <tex@i18nguy.com>
Date: Sun, 19 Oct 2003 15:52:08 -0400
Message-ID: <3F92EB68.BE10DBB6@i18nguy.com>
To: W3C Style <www-style@w3.org>, W3c I18n Group <w3c-i18n-ig@w3.org>

http://www.w3.org/TR/CSS21/visuren.html#direction

Says:

"If a document contains right-to-left characters, and if the user agent
displays these characters in right-to-left order, the user agent must apply the
bidirectional algorithm."


I wonder if this requirement statement is precise and comprehensive.

1) Perhaps it should state what the bidi algorithm should be applied to? The
whole document? The smallest block containing all the right-to-left characters?
Only blocks containing rtl characters?

I can imagine implementors trying to reduce the use of the bidi algorithm to
maximize performance, and perhaps not identifying the impact correctly. There
is an impact due to neutral characters being reordered, so just identifying
right-to-left characters is not sufficient.

2) Since direction can be set in CSS and direction has some impact on the
document even if there are no RTL characters, perhaps the requirement should
reference this possibility. For example table layout would be affected.

So the requirement should perhaps read:

"If the user agent displays right-to-left characters in right-to-left order,
and
if a document contains right-to-left characters or has any right-to-left
direction settings (in Markup or via CSS), 
the user agent must apply the bidirectional algorithm to the document."

(I know this needs tweaking but thought I would throw it out to clarify my
comments.)

tex


-- 
-------------------------------------------------------------
Tex Texin   cell: +1 781 789 1898   mailto:Tex@XenCraft.com
Xen Master                          http://www.i18nGuy.com
                         
XenCraft		            http://www.XenCraft.com
Making e-Business Work Around the World
-------------------------------------------------------------
Received on Sunday, 19 October 2003 15:52:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:24 GMT