W3C home > Mailing lists > Public > www-international@w3.org > January to March 1998

Re: International numbering styles

From: Martin J. Duerst <duerst@w3.org>
Date: Thu, 22 Jan 1998 19:44:00 +0900
Message-Id: <199801221042.TAA15883@sh.w3.mag.keio.ac.jp>
To: Bert Bos <Bert.Bos@sophia.inria.fr>, www-international@w3.org
At 11:39 98/01/09 +0100, Bert Bos wrote:
> The CSS2 specification needs a way to specify numbering styles for
> lists and other things, that goes beyond the small set that CSS1
> provides. I'm trying to find out how people number things in different
> languages, how much of that we need in CSS, and finally, how to refer
> to the numbering schemes in the CSS language.


You gave a lot of interesting examples. Some of them are frequently
used, others are rather rare. I think the basic problem with CSS
numbering styles now is that they only care for the Latin-based
world. Therefore, for CSS2, we should concentrate on adding a
series of keywords that give people in other parts of the world
a chance to do something more local.

> CSS1 has the following:
> 
> 'decimal'            1, 2, 3, 4,...
> 'lower-alpha'        a, b, c, d,...
> 'upper-alpha'        A, B, C, D,...
> 'lower-roman'        i, ii, iii, iv,...
> 'upper-roman'        I, II, III, IV,...
  
I suggest that we introduce

  'lower-latin'        (alias to 'lower-alpha')
and
  'upper-latin'        (alias to 'upper-alpha')
because in an international environment, 'alpha' is quite
ambiguous. The same applies to 'decimal'. If somebody has
a good idea on how to call it, please tell me.


Then we should add a list that could look as follows:
(see at the end of the list for comments)


  'lower-greek'        alpha, beta, gamma,...
  'upper-greek'        Alpha, Beta, Gamma,...
  'lower-greek-old'    similar to Hebrew
  'upper-greek-old'    same as above
    (we need this only if it's in real use currently;
        maybe somebody has a better name)
  'lower-cyrillic'     a, w, b, g,...
  'upper-cyrillic'     A, W, B, G,...

  'lower-armenian'
  'upper-armenian'

* 'hebrew'             as described by Jonathan

  'arabic-letters'
  'arabic-indic-numbers'
  'eastern-arabic-indic-numbers'

  'devanagari-letters'
  'devanagari-numbers'
  'bengali-letters'
  'bengali-numbers'
  'gurmukhi-letters'
  'gurmukhi-numbers'
  'gujarati-letters'
  'gujarati-numbers'
  'orija-letters'
  'orija-numbers'
  'tamil-letters'
  'tamil-numbers'
  'telugu-letters'
  'telugu-numbers'
  'kannada-letters'
  'kannada-numbers'
  'malayalam-letters'
  'malayalam-numbers'

  'thai-letters'
  'thai-numbers'
  'lao-letter'
  'lao-number'
  'tibetan-letters'
  'tibetan-numbers'

  'georgian'

  'hangul'


* 'cjk-ideographic'    plain ideographic numbers
* 'hiragana'           a, i, u, e, o, ka, ki,...
* 'katakana'           A, I, U, E, O, KA, KI,...
* 'hiragana-iroha'     i, ro, ha, hi, ho, he, to,...
* 'katakana-iroha'     I, RO, HA, HI, HO, HE, TO,...

  'bopomofo'


That's what I got from leafing through the Unicode standard.
The items marked with a star have been mentionned as being
used. As we don't want to define things that don't exist,
it would be nice if people could tell us whatever things
they know. We don't want to get from a Latin-based thing
to a Latin+Japanese-based thing.


For CSS3, we can see what we can do with more flexible
ways to specify numbering schemes. One thing we could do
is to allow users to define their own numbering schemes.
The only way that is general and easy enough for CSS that
I can see is to have the numbers enumerated. The problem
with this approach is that you can only use as many numbers
as are enumerated.


Regards,   Martin.
Received on Thursday, 22 January 1998 05:43:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 19:16:52 GMT