W3C home > Mailing lists > Public > www-style@w3.org > October 2013

Re: [css-writing-modes] feedback: bottom-to-top scripts?

From: fantasai <fantasai.lists@inkedblade.net>
Date: Mon, 14 Oct 2013 02:57:51 -0700
Message-ID: <525BC01F.8070803@inkedblade.net>
To: mholmes@uvic.ca
CC: www-style@w3.org
On 10/09/2013 11:11 AM, Tab Atkins Jr. wrote:> On Tue, Sep 3, 2013 at 10:52 AM, Martin Holmes <mholmes@uvic.ca> wrote:
>> Hi there,
>> The current draft of the CSS Writing Modes specification states:
>> "Inherently bottom-to-top scripts are not handled in this version. See
>> [UTN22] for an explanation of relevant issues."
>> However, I can't find in UTN 22 any justification for the omission of
>> bottom-to-top scripts; UTN 22 in fact includes:
>> ltr-btt
>>          Left to right directionality in horizontal text; Bottom to top
>>          directionality  in vertical text. (Bi-orientational script)
>>          Example: Ogham.
>> Other bottom-to-top scripts exist (such as Ancient Berber).
>> I'm therefore a little puzzled by the decision to exclude bottom-to-top from
>> CSS Writing Modes. Is it simply because the range of use-cases is so narrow,
>> so the implementation burden placed on user-agent developers was not felt to
>> be justifiable?
> You got it in one.

No, that is not the reason. The reason is that bottom-to-top scripts
are also left-to-right scripts. However, bottom-to-top in mixed
orientation modes is the *opposite* direction of other LTR scripts
(such as Chinese). Which puts two different LTR scripts going in
opposite directions on the same line. As UTN22 explains, this requires
essentially treating the bottom-to-top script as RTL to get implicit
reordering correct and doing some fancy orientation work to get the
glyphs facing the right way.

The way to handle bottom-to-top scripts in CSS3 is to assign them
'text-orientation: sideways-left'. This deals with all those issues
at a layout and markup level rather than involving a modified bidi

(I suppose that sentence should probably be modified to say that
they aren't handled in any *automatic* way in 'mixed' orientation
but they can be handled by explicitly assigning the correct

Let me know if that makes sense. If not, I'll try to explain it
better. :)

Received on Monday, 14 October 2013 09:58:25 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:51:03 UTC