W3C home > Mailing lists > Public > www-style@w3.org > May 2013

[css3-fonts] font-language-override: substantive comments

From: fantasai <fantasai.lists@inkedblade.net>
Date: Thu, 23 May 2013 20:05:33 +0800
Message-ID: <519E060D.1020706@inkedblade.net>
To: "www-style@w3.org" <www-style@w3.org>
http://dev.w3.org/csswg/css-fonts/#font-language-override-prop

   # The value of ‘normal’ implies that when rendering with
   # OpenType fonts the language of the document is used to infer
   # the OpenType language system, used to select language specific
   # features when rendering. The value of the <string> is a single
   # three-letter OpenType language system tag, defined in the
   # layout tag registry of the OpenType specification.

So, 'normal', is defined, but while the syntax of the other
values is defined, the semantics of it is not, except by
example (below). This feature needs an actual normative
definition here.

   # the language of the document is used to infer the OpenType
   # language system

Are we sure we want the language of the document, and not
the language of the element, here?

If so, let's highlight this distinction, as I'm sure it will
be missed by many.

   # However, a given font may lack support for a specific
   # language. In this situation authors may need to use the
   # typographic conventions of a related language that are
   # supported by that font:
   #
   # <body lang="mk">     <!-- Macedonian lang code -->
   # body { font-language-override: "SRB";
   #        /* Serbian OpenType language tag */ }

If this is actually a use case we want to support, I'd like to
see this property take a comma-separated list of values, so that
if the font actually used does support the correct language, it
can be activated. (Suppose for example we get a fallback font,
or the font gets updated to handle Serbian.)

Also the example here should follow good practice and select
off of 'lang'.

   [lang=mk] { font-language-override: "MKD", "SRB"; }

Lastly, are these codes case-sensitive? We should point that
out, one way or the other.

~fantasai
Received on Thursday, 23 May 2013 12:06:07 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 23 May 2013 12:06:08 UTC