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

Re: [css3-lists] Proposal for a generic numeric list-style-type

From: Markus Ernst <derernst@gmx.ch>
Date: Tue, 18 May 2010 10:55:17 +0200
Cc: www-style@w3.org, gabriele.romanato@gmail.com
Message-ID: <20100518085517.318840@gmx.net>
To: timeless <timeless@gmail.com>, Simetrical+w3c@gmail.com
-------- Original-Nachricht --------
> Datum: Tue, 18 May 2010 11:09:41 +0300
> Von: timeless <timeless@gmail.com>

> On Tue, May 18, 2010 at 2:34 AM, Aryeh Gregor <Simetrical+w3c@gmail.com>
> wrote:
> > I mean, are there any languages where <ol> will display by default
> > with localized, non-decimal numerals?
> 
> Well, if you consider java implemented browsers:
> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4210199
> 
> http://www-archive.mozilla.org/releases/mozilla1.8b1/known-issues-int.html
> The default for contextual numeric alteration has changed since
> Mozilla 1.4 Western digits in Arabic documents are no longer
> automatically replaced with Arabic digits  in a context-sensitive
> manner. To activate the contextual replacement of Western digits with
> Arabic ones, type about:config  in the location bar and set the value
> of the preference item bidi.numeral to 1 (Bug 181711). Alternatively,
> you can edit user.js by locating the file in the profile directory and
> using a text editor that can handle UTF-8 encoding.
> 
> https://bugzilla.mozilla.org/show_bug.cgi?id=181711
> https://bugzilla.mozilla.org/show_bug.cgi?id=151374
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/intl/nls_0to3.asp
> http://www.unicode.org/book/ch13.pdf, p. 320
> 
> > By default, we should always display the nominal forms with no
> substitution.
> 
> http://lists.w3.org/Archives/Public/public-iri/2003May/0010.html
> 
> http://msdn.microsoft.com/en-us/library/dd317791(VS.85).aspx
> http://msdn.microsoft.com/en-us/library/dd373769(VS.85).aspx
> 
> http://blogs.msdn.com/michkap/archive/2005/09/09/462920.aspx
> http://blogs.msdn.com/michkap/archive/2004/12/01/272864.aspx
> 
> Probably one of my favorite entries on this topic is:
> http://blogs.msdn.com/michkap/archive/2006/10/02/783066.aspx
> 
> I think the short answer is "yes, there are some Windows locales where
> '0' .. '9' are not rendered as Western, and thus there should be some
> versions of at least IE for which this is the case."

That's interesting, though I don't understand all of it. Before I read the CSS3 lists spec, as a dumb, euro-centric web author, my expectation was that the list item markers in an <ol> by default would adapt to what is most common in the element's language (not the user language), respectively to what browser implementors assume to be most common. I would have expected list item markers to be different in both lists of a code such as:

<html lang="de">
  <body>
    <ol>
      <li>...</li>
    </ol>
    <ol lang="ja">
      <li>...</li>
    </ol>
  </body>
</html>

To achieve this with the current CSS3 spec I need to specify the list-style-type value separately for all languages used. Even more, I need to know the exact HTML structure when writing the CSS, as language attributes can occur in any of the parent elements of the list. So, from an author's POV a generic numeric value seems highly desirable to me.

I even still think the generic value would be a good default, as the most common case should not require extra code. Authors aware of local or contxtual variations are still free to use their preferred list-style-types (they have to do this with the current spec anyway).

Of course, after reading Aryeh Gregors messages I am aware of the fact that I have little idea about possible difficulties in defining what is most common in some languages. Inconsistent rendering across browsers could occur. I am not sure whether that would be a big problem in this case, as authors who really care can just apply the list-style-type property.
-- 
GMX.ch - Schweizer FreeMail-Dienst mit ├╝ber 800.000 Mitgliedern
E-Mail & mehr! Kostenlos: http://portal.gmx.net/de/go/chfreemail
Received on Tuesday, 18 May 2010 09:22:32 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:27 GMT