- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Mon, 13 Sep 2010 17:48:32 -0700
- To: John Hudson <tiro@tiro.com>
- Cc: Sergey Malkin <sergeym@microsoft.com>, John Daggett <jdaggett@mozilla.com>, Sylvain Galineau <sylvaing@microsoft.com>, www-style@w3.org, www-font <www-font@w3.org>
On Mon, Sep 13, 2010 at 5:32 PM, John Hudson <tiro@tiro.com> wrote: > Sergey Malkin wrote: > >> This is what bothers me. This answer means simulated styles will never be >> used even if just single font is defined (like MyFont1 above). I do not >> think this is what Web developers would expect. This is different from >> people's experience with fonts installed locally... > > If I understand you correctly, I think I would want to nuance this by saying > 'different from people's experience with fonts installed locally in some > applications'. Professional design apps such as Adobe's do not employ > simulated styles except when explicitly activated by the user. In such apps, > the absence of an Italic font means no italic, not a simulated italic, and > in my opinion as a typographer that is vastly preferable to what apps like > Word do, mutilating typefaces in numerous ways with simulated styles, even > for single fonts that were never meant to be italic'd or bold'd. In my opinion as a simple web author, though, the exact opposite is true. ^_^ I'd much rather have a simulated font if there's no appropriate variant specified, rather than just not matching at all and falling back. For example, check out http://www.xanthir.com/:wih in Firefox and Chrome. The former will simulate font-variant:small-caps for my headings, which looks fine. The latter won't, so the headings fallback to the platform serif. I'd prefer either simulation happening automatically, or at the very least a switch saying that it's okay to simulate some/all properties that aren't otherwise matched by an explicit declaration. Basically, I'm not at all sympathetic to a typographer/font developer saying "I don't want my font used at all if it's used in a way I can't control the display of", which I believe is essentially the argument of the no-simulation camp. (Correct me if there is a more nuanced position I should be aware of.) ~TJ
Received on Tuesday, 14 September 2010 00:49:25 UTC