W3C home > Mailing lists > Public > www-style@w3.org > August 2015

Re: [css-writing-modes] Propose writing-mode: sideways-left

From: Hiroshi Sakakibara <sakakibara.hiroshi@bpsinc.jp>
Date: Fri, 21 Aug 2015 17:50:22 +0900
To: Florian Rivoal <florian@rivoal.net>
Cc: John Daggett <jdaggett@mozilla.com>, Jonathan Kew <jfkthame@gmail.com>, Koji Ishii <kojiishi@gmail.com>, fantasai <fantasai.lists@inkedblade.net>, "www-style@w3.org" <www-style@w3.org>
Message-Id: <20150821175021.DD36.147D1882@bpsinc.jp>
> This proposal was made in part due to the reluctance of implementors to implement sideways-left, so yes, it would be very interesting to hear what they think about this new one.

My company's developer says, the implementation becomes easy.

But since we haven't met the situation to use the sideways-left, the
priority to develop it is not so much high. 

(For example if we need to write sideways-left in table, it is enough
to use transform, or something like that. The situation is rare.)

--- skk 

On Fri, 21 Aug 2015 10:13:11 +0200
Florian Rivoal <florian@rivoal.net> wrote:

> 
> > On 21 Aug 2015, at 08:32, John Daggett <jdaggett@mozilla.com> wrote:
> > 
> > 
> > Jonathan Kew wrote:
> > 
> > > I don't agree that the proposal being considered here would increase
> > > authoring model complexity. If anything, I'd say it offers authors a
> > > cleaner and more understandable model. We'd have three modes
> > > (horizontal-tb, sideways-lr and sideways-rl) that all lay out text in
> > > the same way, but with a ±90° rotation in the sideways-* cases. In all
> > > three cases, the text is laid out according to the conventions of
> > > horizontal writing, even if it is then rotated in its entirety. No
> > > question of glyph orientation within the line ever arises in these
> > > modes.
> > 
> > I guess it boils down to this for me: for 99% of the users of the
> > 'writing-mode' property, namely authors in Japan, Taiwan and Hong Kong
> > laying out vertical text runs, the sideways-* values will be a confusing
> > addition.
> 
> CJK users of the writing-mode property can ignore these new values, as they
> are here to server other uses cases. If they do ignore them, the model is not
> changed for them
> 
> horizontal-script users of the writing-mode property, on the other hand, will
> find it much easier to use, since it now has values that directly address their
> use cases.
> 
> > I do understand that this simplifies implementations details. You no
> > longer need to be concerned about mixtures of sideways-left and
> > sideways-right text in the same block. I just am concerned that we're
> > adding values to a commonly-used property, 'writing-mode', for the sake
> > of simplifying a less commonly used property, 'text-orientation'.
> 
> It not only simplifies the implementation of text-orientation, it also simplifies
> usage, as it is less commonly needed than before.
> 
> > Florian Rivoal wrote:
> > 
> > > > I'd guess that for CJK authors, the use of text-orientation:sideways
> > > > will be virtually non-existent, as that simply isn't how these
> > > > languages are written vertically.
> > > 
> > > text-orientation:sideways should be used by CJK authors to market small piece of foreign language embedded in a their text.
> > > 
> > > article {
> > >   writing-mode: vertical-rl;
> > >   text-orientation: upright;
> > > }
> > > 
> > > article q:lang(en) {
> > >   text-orientation: sideways.
> > > }
> > 
> > I think both of these statements are off the mark. The initial value of
> > 'text-orientation' is 'mixed', such that the contents of vertical text
> > elements will be laid out based on UTR-50 default orientation data. So
> > the natural default for vertical text is simply:
> > 
> > article {
> >   writing-mode: vertical-rl;
> >   /* default text-orientation: kana/kanji are upright, latin is sideways */
> > }
> > 
> > No need to use text-orientation or extra markup for small runs of Latin
> > text.
> 
> "text-orientation: mixed" is a heuristic. It will most of the time, but importantly not always, do the right thing, as some punctuation characters can go both ways. If an CJK author knows their text is pure CJK, they should use text-orientation: upright. If they know that it is pure CJK, except for small runs of latin (or other horizontal script) text which are appropriately marked up, they should use upright on the CJK text, and sideways on the small runs. If they don't know, or if they know that the small runs are not marked up in a distinctive way, text-orientation: mixed will probably do a good enough job, but not a perfect one.
> 
> All this is the same under the old and the new proposition.
> 
> 
> On the other hand, the old design required that the text-orientation property with the sideways-* values also be used for horizontal script authors. The situation is simplified, as they can now ignore this property when using the appropriate writing modes.
> 
> > I really think we need more input from other implementers, specifically
> > Apple, Microsoft and any other EPUB vendors who support already support
> > vertical text. Murakami-san seems content with the proposed change.
> > Other implementers, opinions?
> 
> This proposal was made in part due to the reluctance of implementors to implement sideways-left, so yes, it would be very interesting to hear what they think about this new one.
> 
>  - Florian


--
榊原 寛 / Sakakibara Hiroshi (sakakibara.hiroshi@bpsinc.jp)
ビヨンド・パースペクティブ・ソリューションズ株式会社
〒160-0023 東京都新宿区西新宿6-20-7
コンシェリア西新宿TOWER'S WEST 2F
Tel: 03-6279-4320 Fax: 03-6279-4450
Web: http://www.bpsinc.jp/
Received on Friday, 21 August 2015 08:52:06 UTC

This archive was generated by hypermail 2.3.1 : Friday, 21 August 2015 08:52:07 UTC