W3C home > Mailing lists > Public > www-style@w3.org > March 2011

[css3-fonts] Variants (Ligatures, Capitalization, Digits)

From: Christoph Päper <christoph.paeper@crissov.de>
Date: Wed, 9 Mar 2011 18:02:33 +0100
Message-Id: <D4A51C7F-59C3-44F1-90D5-1EED2DDD1B23@crissov.de>
To: www-style list <www-style@w3.org>
Thanks for incorporating some of my earlier notes and suggestions.

John Daggett:
> * Section 6.4 subscript, superscript, ordinal description

This actually might work.

* Section 6.5 Ligatures

“nums” matches “caps”, so why not “ligs”? :)

People will probably want to use

  font-variant-ligatures: none;

and

  font-variant-ligatures: all;

For defaults specified by the font ‘auto’ seems more natural than ‘normal’, but the latter is usually used for shorthand resets.

I have said before <http://lists.w3.org/Archives/Public/www-style/2010Mar/0049.html> that I don’t think the current design of this feature is sound and we should rather use something user-friendly like

  font-variant-ligatures: all | additional | common | none | auto/normal 

and leave users with special needs with ‘font-feature-setting’. 

You would, at first glance, need something more verbose to make ‘font-variant’ being a shorthand possible (6.11), at least if you wanted to be able to set everything within the shorthand. At second thought, ligatures, capitals and digits don’t share keywords, even if you leave out “-ligatures”, “-caps” and “-nums”.

* Section 6.6 Capitalization

I have outlined my concerns earlier. This time I only want to suggest a change from

| <caps-value> = small-caps | all-small-caps | petite-caps | all-petite-caps
|              | titling-caps | unicase

to

  <caps-value> =  all? [ small-caps | petite-caps | titling-caps ]
               | unicase

If ‘all’ is specified with ‘small-caps’ or ‘petite-caps’, OT feature ‘c2sc’/‘c2pc’ is applied and if that is impossible, as a fallback, ‘text-transform’ is set to ‘lowercase’.
If ‘all’ is specified with ‘titling-caps’, ‘text-transform’ is set to ‘uppercase’ (since there is no corresponding OT feature).

If ‘text-transform’ is set to ‘uppercase’, ‘font-variant-caps’ is reset in a way that the OT feature ‘titl’ becomes activated.

* Section 6.7 Numerical formatting

I also still believe ‘font-variant-caps’ and ‘text-transform’ should interact with ‘font-variant-numeric’, i.e. <numeric-figure-values>:
If ‘text-transform’ is set to ‘uppercase’ (directly or indirectly), ‘font-variant-numeric’ is reset in a way that the OT feature ‘lnum’ becomes activated.
If ‘text-transform’ is set to ‘lowercase’ (directly or indirectly), ‘font-variant-numeric’ is reset in a way that the OT feature ‘onum’ becomes activated.

And I still believe the value names are not well chosen, and neither are they in Open Type:

  “lining”       (+lnum, -onum)
  “casing”       (+onum, -lnum)
  “proportional” (+pnum, -tnum)
  “monospaced”   (+tnum, -pnum)

  “oldstyle”     (+onum, +pnum)
  “tabular”      (+tnum, +lnum)
   …

* Section 6.10 East Asian text redering

This section is about sinograms (Hanze, Kanji, Hanja …), not about all scripts of East Asia. I argued before that the property described herein (partially) could be used for other scripts in the future and should therefore (be designed differently or at least) get a more neutral name, please consider ‘font-variant-sinograms’.

The meaning of <integer> for ‘simplified’ (OT ‘smpl’) and ‘traditional’ (‘trad’) is not explained and doesn’t seem strictly necessary at first glance.

* Section 6.13 Font language override

Unlike implementors, authors cannot be burdened with looking up language codes in different places, so use BCP47 instead of OT language codes.
Received on Wednesday, 9 March 2011 17:03:07 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:38 GMT