W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2004

Re: erratum? Re: 04 Feb 2004 - Techniques

From: Charles McCathieNevile <charles@w3.org>
Date: Fri, 17 Sep 2004 08:27:19 -0400 (EDT)
To: Joe Clark <joeclark@joeclark.org>
Cc: WAI-GL <w3c-wai-gl@w3.org>
Message-ID: <Pine.LNX.4.55.0402050405230.9108@homer.w3.org>

On Wed, 4 Feb 2004, Joe Clark wrote:

>I can't believe I read something from Chaals.

Oh. I thought you were bound by the same conditions I am in participating in
this list - which includes "remain current on w3c-wai-gl list and respond in
timely manner to postings", so assumed you read at least some stuff that I

>> According to the CSS specification, as Joe points out, px is a relative unit.
>And that settles the issue of relative vs. absolute units. They are
>defined by the W3C, of which WAI is a part.

And as a part, can ask the W3C to change the definition offered by CSS. (Have
a look at the cascading rules in CSS1 and CSS2 for a precedent where this

Actually the point is that the two groups have used "relative" without
qualifying what something is relative to, which makes it confusing.

>> However, unlike the others, it is relative to the user's screen, and not
>> so readily adjusted - instead of changing font size you have to change your
>> display settings (or your screen), which I think is more complicated for most
>> users.
>> As one of the originators of this checkpoint I certainly meant that, and
>> apologise for having been a bit sloppy 5 or 6 years ago.
["That" refers to units relative to the font-size or window size - things
readily adjusted by the user even in real time]

In order to overcome the sloppiness, I was (and still am) proposing an
erratum to clarify a spec that a lot of the world are actually using and
trying to interpret, so it is clear what was meant.

The question of whether px, as a unit that is relative to either a hardware
setting or in some cases a software setting, should be considered an
appropriate unit for specifying sizes of text or layouts is one for a spec
that replaces WCAG 1. Clarifying an editorial error in the existing
specification so the world is clearer about what it got seemed to me like an
obviously easy and useful thing to do. Most W3C groups do it from time to
time (HTML, CSS, XML, MathML...)


Received on Friday, 17 September 2004 12:27:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:51 UTC