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

Re: [css-writing-modes][CSS21] propagation of 'direction' from <body>

From: Koji Ishii <kojiishi@gmail.com>
Date: Sat, 22 Aug 2015 01:11:23 +0900
Message-ID: <CAN9ydbVxcQjFNe44OYjXO37dkKP=5g6-0gMMOUpSODE98pSQ4Q@mail.gmail.com>
To: Shinyu Murakami <murakami@vivliostyle.com>
Cc: Simon Pieters <simonp@opera.com>, Greg Whitworth <gwhit@microsoft.com>, fantasai <fantasai.lists@inkedblade.net>, "www-style@w3.org" <www-style@w3.org>
I ran some investigation on this topic, both internal and external. My
investigation is primarily on writing-mode.

In short, I agree with Simon that we should propagate both writing-mode and
direction from body. I understand not to propagate is slightly simpler, but
the risk looks too high to me, higher than the benefits of not to propagate.

Some information follow:

1. We wanted to look into EPUB content, but the most common pattern in EPUB
is to use classes in HTML and define class selectors in CSS, so the
investigation is quite more than grep. It's technically possible, but we do
not have the bandwidth to do it at this moment.

2. From my experiences working on EPUB content issues before, there were
such content. I wasn't counting, but the rate was not low.

3. Most EPUB applications do more than just loading HTML files into UA.
After loading content along with app's fragments and JS, apps calculate
writing-mode by itself and set to UA, so changing "not to propagate" could
break apps. I talked to two developers, one developer said for sure the
change breaks his app. The other could not answer for sure but said likely
to break.

Sorry for not being able to get the concrete data in time, but even without
this, the data on the web from Simon indicates me that the risk is way too
high.

/koji
Received on Friday, 21 August 2015 16:12:12 UTC

This archive was generated by hypermail 2.3.1 : Friday, 21 August 2015 16:12:12 UTC