W3C home > Mailing lists > Public > www-style@w3.org > January 2008

Re: Advanced Font Features

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Sat, 19 Jan 2008 15:30:15 +0200
Message-ID: <017e01c85a9f$6db6de20$0500000a@DOCENDO>
To: "CSS Style" <www-style@w3.org>

Christoph Päper wrote:

> Font technologies like Open Type and AAT provide a number of advanced
> typographic features, i.e. stuff known from print or handwriting, but
> uncommon to electronic typesetting. Not all of them have counterparts
> in CSS yet.
>
> Should they? How?

That would be a job for CSS, but I wonder how realistic the idea is. I 
would like to see decent CSS 2.0 (or at least CSS 2.1) support in 
browsers first, then some hyphenation "control", before considering 
ligation

>    code    {ligation: none;}

Something like that would be nice indeed, though probably ligation: none 
should be the default.

As the first step, we could add a property with a list of strings as its 
value, specifying the letter combinations that should be ligated, e.g.
body { ligation: "fi", "ffi"; }

>    table   {number-style: tabular non-lining;}

Sounds fine, though usually we can and should deal with numeric tables 
by selecting a font that has the desired kind of digit glyphs. Switching 
between "old style digits" and "uppercase digits" within text would be 
more relevant, since in text, you might wish to use old style digits as 
a rule but uppercase digits in expressions involving uppercase letters 
(e.g., "F-1" or "ISO 10646" does not look very good when old style 
digits are used). This however would imply that you need extra markup 
(<span>, probably) to indicate the parts that should use uppercase 
digits.

Additionally, it would be nice to be able to affect kerning, perhaps 
just as a simple on/off option, perhaps with values like none, metric, 
optic, or even with an option of specifying an explicit kerning table.

> PS: As far as I know there is nothing like italics in East-Asian
> typography.

Italics is limited to a few scripts.

> Oblique can be simulated by automatic slanting.

But that's usually a bad idea.

Returning to the practical situation, I wouldn't suggest taking much 
trouble in defining new cool features before browsers can at least 
distinguish between italics and oblique and stop treating them as the 
same. Specifically, font-style: italic should be ignored when the font 
has no italics version.

> Would it
> be acceptable for Japanese to switch display between Katakana and
> Hiragana for "font-style: italic"?

Hardly.

Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/ 
Received on Saturday, 19 January 2008 13:30:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:58 GMT