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

Re: [css3-fonts] @font-face matching and font-style descriptor

From: John Daggett <jdaggett@mozilla.com>
Date: Wed, 15 Sep 2010 12:09:52 -0700 (PDT)
To: Sergey Malkin <sergeym@microsoft.com>
Cc: "Tab Atkins Jr." <jackalmage@gmail.com>, John Hudson <tiro@tiro.com>, Sylvain Galineau <sylvaing@microsoft.com>, www-style@w3.org, www-font <www-font@w3.org>, Thomas Phinney <tphinney@cal.berkeley.edu>, Brad Kemper <brad.kemper@gmail.com>
Message-ID: <1489444064.580562.1284577792754.JavaMail.root@cm-mail03.mozilla.org>

Sergey Malkin wrote:

> > The problem with allowing "fallback to a font family that has real
> > italics" is that it doesn't work for scripts that typically lack
> > italics (e.g. Japanese).
> 
> This is not problem of CSS Fonts module at all . Keyword here is
> 'script'. Do you homework. Understand what is used for emphasis for
> language used on the page and  _if_ you need italic , use italic
> font. Only now you decide which font it will be.

No, it's a *spec* problem because there's no font to fallback to in
the case of Japanese if fallback is defined this way.  I'm perfectly
aware that Japanese doesn't have an italic tradition.  But users in
Japan now expect that Japanese text in <em> tags to be obliqued, so
disabling synthetic italics is problematic.

Not sure what homework I need to do. ;)

> > If such option is introduced, I see need for three values: allow
> > synthetic styles, use specified face without faux style, or
> > fallback
> 
> I started this sentence with "if", which does not mean I support
> existence of such option. This only means that if it exists, it has
> to include all three options. I prefer Web designer to pay attention
> to what is going on with site being developed. If this is done,
> situations with unintended simulated styles will be more like
> emergency fallback not worse than accidental fallback to system
> fonts because of unexpected characters. Current spec gives you power
> to implement that.

Agreed, I wasn't advocating for those properties, I was simply trying
to describe properties that would cover the behavior John Hudson
originally noted.

Fallback behavior is always less than ideal, introducing complexity to
deal with this isn't necessarily the right solution.

Regards,

John Daggett
Received on Wednesday, 15 September 2010 19:10:26 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:31 GMT