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

Re: [CSS21] computed value of 'font-weight' is not precisely defined

From: John Daggett <jdaggett@mozilla.com>
Date: Tue, 10 Jun 2008 23:35:49 -0700 (PDT)
To: "L. David Baron" <dbaron@dbaron.org>
Cc: www-style@w3.org
Message-ID: <10880553.380731213166149854.JavaMail.root@cm-mail02.mozilla.org>


> (1) the computed value is a legal number value combined with an
>    <em>ordered sequence</em> of relative values.
> 
> (2) the computed value is a legal number value combined with a count
>    of relative values (with positive vs. negative distinguishing
>    bolder and lighter)

Both of these ways have merits but maintaining an ordered sequence of bolder/lighter steps is a big implementation headache, the sequence is
essentially bounded only by the element nesting level.  I don't see any strong advantage to it and the added complexity is a huge disadvantage.

Given a font family with two weights, 400 and 700, consider the following example:

p { font-weight: 700; }
strong { font-weight: bolder; }
em { font-weight: lighter; }

<p>Music can be <strong>loud or <em>soft</em></strong> yet soothing</p>

(1) with an ordered sequence:

loud is bolder than soft
soft is lighter than soothing

(2) with relative delta weights:

loud is the same weight as soft
soft is the same weight as soothing

The first way satisfies the author who says "when I use lighter I want lighter dammit" but annoys the author who wonders "why is text in a strong tag lighter than text outside that tag?".

I think we should change the wording to describe how computed weights are calculated and how a face is selected from that value. 

John Daggett
Mozilla Japan
Received on Wednesday, 11 June 2008 06:36:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:07 GMT