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

RE: [css3-writing-modes] a third option for implementing logical properties

From: Koji Ishii <kojiishi@gluesoft.co.jp>
Date: Mon, 25 Oct 2010 01:28:47 -0400
To: John Daggett <jdaggett@mozilla.com>, "MURATA Makoto (FAMILY Given)" <eb2m-mrt@asahi-net.or.jp>
CC: "www-style@w3.org" <www-style@w3.org>
Message-ID: <A592E245B36A8949BDB0A302B375FB4E0AA3634143@MAILR001.mail.lan>
You're right. But it doesn't deny the value of logical directions.

We must add something to support vertical flow. Without logical directions, you may end up adding one for each combination of writing-mode and direction, so it could go up to 8. With logical directions, you add just one.

Also there's a hope that, not guaranteed but, in 10 years or so, we could stop doing that. Wrapping CSS with <!-- and --> was the best practice 10 years ago, and we no longer do that. Without logical directions, we don't have that hope either. You have to keep writing 2 to 8 styles per direction.


-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf Of John Daggett
Sent: Monday, October 25, 2010 2:10 PM
To: MURATA Makoto (FAMILY Given)
Cc: www-style@w3.org
Subject: Re: [css3-writing-modes] a third option for implementing logical properties

Makoto Murata wrote:

> Murakami-san,
> Thank you for providing real examples.  Your horiz.css and vert.css 
> are very helpful.  I would recommend everybody to use Murakami-san's 
> example for demonstrating advantages of your favorite options.
> http://nadita.com/murakami/test-logical/vert.css

> http://nadita.com/murakami/test-logical/horiz.css

> Here I give a unified diff result.  Obviously, maintaining these two 
> sets is too painful.  Note that automatic conversion is not always 
> possible since the stylesheet author might want to introduce some 
> differences on purpose.

But to support fallback in older browsers, the version of this stylesheet used if the proposed logical properties are introduced isn't much better, all asymmetric property values need to have a fallback form and a logical form:

  html {
    font-size: 12pt;
    line-height: 1.75;
  div.header {
    border-top: thick solid;
    border-before: thick solid;
    border-bottom: thick solid;
    border-after: thick solid;
    margin-bottom: 2em;
    margin-after: 2em;
    line-height: 1.5;
  h1.title {
    font-size: 3em;
    padding-left: 1em;
    padding-start: 1em;
    margin: 0;
  h2.author {
    text-align: right;
    text-align: start;
    font-size: 2em;
    padding-right: 1em;
    padding-end: 1em;
    margin: 0;
  div.article {
    width: 70%;
    padding-right: .5em;
    padding-end: .5em;
    border-right: solid thin;
    border-end: solid thin;
    text-align: justify;
  div.aside {
    width: 26%;
    logical-width: 26%;
    float: right;
    float: end;
    padding-left: .25em;
    padding-start: .25em;
    padding-right: .25em;
    padding-end: .25em;
    background-color: #cdf;
  p {
    margin: 0;
    text-indent: 1em;
  blockquote {
    margin-left: 2.75em;
    margin-start: 2.75em;
    margin-right: 2.75em;
    margin-end: 2.75em;
    margin-top: .25em;
    margin-before: .25em;
    margin-bottom: .25em;
    margin-after: .25em;
    padding: .25em;
    background-color: #fd8;

John Daggett

Received on Monday, 25 October 2010 05:26:50 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:40 UTC