W3C home > Mailing lists > Public > www-style@w3.org > October 2015

Re: [css-fonts] font-weight-adjust

From: Florian Rivoal <florian@rivoal.net>
Date: Fri, 23 Oct 2015 16:48:44 +0900
Cc: Jonathan Kew <jfkthame@gmail.com>, www-style <www-style@w3.org>
Message-Id: <A3524F1F-2CED-452F-8281-86CB929D268B@rivoal.net>
To: John Daggett <jdaggett@mozilla.com>

> On 23 Oct 2015, at 16:10, John Daggett <jdaggett@mozilla.com> wrote:
> 
> Um, you were asking a theoretical question about TrueType collections, which are not supported as a webfont format currently. I don't think there's a problem that the use of a particular font packaging format will necessarily solve.

I was not explicitly asking about TrueType collections, I was asking if you can use fragment identifier to select a single font face out of a font file, regardless of the format of the font file. There are certainly today font files that are supported and that contain multiple weights.

> > So just to clarify:
> > 
> > * Do you agree what allowing authors to pair fonts based on
> > numerically different but visually similar font weights is a desirable
> > thing?
> > 
> > * Do you think the currently-specified-but-not-implemented solution
> > based on fragment identifiers will solve the problem?
> > 
> > * If yes, what are the road blocks to getting it implemented?
> > 
> > * If no, do you think something like font-weigth-adjust has a better
> > chance, and if not, any idea about what would?
> 
> Your original message was about matching fallback fonts. I think your proposed solution is a poor way to solve *that* problem. Decisions about which font to use for a particular language are much subtler than simply being a decision about matching font weights. Type designers spend a lot of time thinking about these sorts of problems when designing typefaces that are complementary across scripts. Using fonts that are explicitly *designed* to complement each other across script/language is a better solution.

It is absolutely a better solution when you can use it, but it may not always be possible. You do not always have a set of complementary typefaces that work well across all the languages/scripts that you want to support, nor the resources to commission someone to make one for you (especially when the missing one has a massing amount of code points to cover, as in the CJK case).

If CJK support is your primary goal, it may not be a huge issue to find a latin font that matches your CJK font, but the other way around is more tricky. So if you're set on your non-CJK choice of fonts, and are trying to extend support to CJK as well, finding a font that fits with the rest is not necessarily simple.

When you don't have a set of perfectly matching typefaces, but you have some that are close enough except they are poorly calibrated in terms of x-height, you can help yourself with font-size-adjust.

I do not really think it matters whether the fallback gets used because the main font wasn't there at all, or because it was but did not support certain unicode ranges. font-size-adjust deals with both.

I'm looking for an equivalent mechanism to help yourself in terms of weight.

> Type is typically used in some form of hierarchy within a page. For example, headings using a "branded" display face, body text using a commonly available platform font. Sites that display content in different languages need to make font choices per language to maintain a consistent typographic voice across locales. Universal fontlists are a poor way of addressing that.
> 
> In short, I don't think your 'font-weight-adjust' proposal is a good or needed solution.

I'm perfectly willing to admit it is a poor solution, but I don't quite see why this is any less useful to solve than text-size-adjust.

 - Florian
Received on Friday, 23 October 2015 07:49:17 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:57 UTC