W3C home > Mailing lists > Public > wai-wcag-editor@w3.org > April to June 2005

Request modification for WCAG 1.0 Errata

From: Roberto Scano - IWA/HWG <rscano@iwa-italy.org>
Date: Thu, 19 May 2005 09:48:28 +0200
Message-ID: <005601c55c47$268fd140$0200a8c0@iwars>
To: <wai-wcag-editor@w3.org>
Cc: "Charles McCathieNevile" <charles@w3.org>

Hi!
In italian mailing list about web accessibility we are discussing about 
the law application and to the reference to WCAG 1.0 checkpoint 3.4.

Use relative rather than absolute units in markup language attribute 
values and style sheet property values. [Priority 2]
For example, in CSS, use 'em' or percentage lengths rather than 'pt' or 
'cm', which are absolute units. If absolute units are used, validate 
that the rendered content is usable.

with its relative techniques:
http://www.w3.org/TR/WCAG10-CSS-TECHS/#units

and the relative errata:
http://www.w3.org/WAI/GL/WAI-WEBCONTENT-ERRATA

Description (and correction). The note for Checkpoint 3.3 should say 
something about the effect of proportional sizing on raster images as 
follows: "Use relative rather than absolute units in markup language 
attribute values and style sheet property values. [Priority 2] For 
example, in CSS, use 'em' or percentage lengths rather than 'pt' or 
'cm', which are absolute units. If absolute units are used, validate 
that the rendered content is usable (refer to the section on 
validation). For example, do not proportionally size raster images."

Also this errata contain a reference to checkpoint 3.3 that should be 
3.4.

There is a problem that is a lot discussed: Pixel. Pixel are relative 
unit of masurement referred to the screen resolution. So, for eg., 
inside a browser the IE font resize is correct because IE resize only 
fonts that have a relative dimension (% or .em, so with a reference to 
the charset default dimension).

Now there are some situation where there is people that said, referring 
to CSS 2.0 Spec (http://www.w3.org/TR/CSS2/syndata.html#length-units) 
that pixel is a relative unit of measurement (and it is!) and so they 
can develop web content (and text) with .px dimension.

This, reading the checkpoint 3.4 and the CSS 2.0 Spec, is true and can 
cause a lot of inaccessible pages that conform (!!!) to WCAG 3.4 
checkpoint and gan guarantee a level "AA" to a website that, for eg, 
cannot autorize the resize of the text.

>From CSS 2.0 spec:
  a.. em: the 'font-size' of the relevant font
  b.. ex: the 'x-height' of the relevant font
  c.. px: pixels, relative to the viewing device

So what I suggest is to modify the actual errata like this:

Description (and correction). The note for Checkpoint 3.4 should say 
something about the effect of proportional sizing on raster images as 
follows: "Use relative rather than absolute units in markup language 
attribute values and style sheet property values. [Priority 2] For 
example, in CSS, use 'em' or percentage lengths rather than 'pt' or 'cm' 
or 'px' (Pixel is relative to the viewing device), which are absolute 
units.
If absolute units are used, validate that the rendered content is usable 
(refer to the section on validation). For example, do not proportionally 
size raster images."

Or find another solution but that clearly define about pixel usage. This 
is important for low-vision people and for the correct law application 
(and for future laws in Europe, due that eEurope project ask to refer to 
WCAG 1.0).

Roberto Scano (rscano@iwa-italy.org)
IWA/HWG International Project Manager and EMEA Coordinator
International Webmasters Association / HTML Writers Guild
W3C Advisory Commitee Representative for IWA/HWG
W3C WCAG Working Group Member - W3C ATAG Working Group Member
Expert of ISO/TC 159/SC 4/WG 5 'Software ergonomics and human-computer 
dialogues'
http://www.iwanet.org - http://www.hwg.org
E-Mail: emea@iwanet.org - w3c-rep@iwanet.org
Personal web site: http://www.robertoscano.info
--------------------------------------------------------------------
Received on Thursday, 19 May 2005 07:48:39 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:04 UTC