W3C home > Mailing lists > Public > public-comments-wcag20@w3.org > August 2005

[Plian Text] IBM comments on WCAG CSS Techniques Working Draft of July 30, 2005

From: <Becky_Gibson@notesdev.ibm.com>
Date: Tue, 23 Aug 2005 10:07:29 -0400
To: public-comments-wcag20@w3.org
Message-ID: <OFAE0535DE.42EA651A-ON85257066.004D8637-85257066.004DE22F@notesdev.ibm.com>

[RESEND as plain text for ease of reading]

There are the general comments from IBM on the WCAG CSS Techniques Working 
Draft of July 30, 2005


General Comments:
Examples should include the source (which is there now) as well as the 
actual example on the page so you can see how it is rendered in your 
browser.
Order the techniques based on the WCAG guideline so they are easier to 
follow.  If that is not possible, then in the Table of Contents include 
the guideline each technique is addressing.
The techniques should clearly indicate the guideline and the level they 
apply to.
There was an editorial note about referencing an HTML technique in the CSS 
techniques document (the one on using the title attribute for labels).  It 
would be a good idea if the techniques were organized by guideline rather 
than by topic area.  Then you could see everything you needed to do for 
Guideline 1.1 for the level you are trying to achieve.

Specific comments:
1.2 Using px for properties that do not need to be changed .  Add a better 
list or description of properties that "do not need to be changed". 
Otherwise it is very subjective.
2.1 Selecting individual characters or lines  Don't see what :first-letter 
or :first-line have to do with accessibility.  Is this saying that these 
pseudo-elements should be used in place of using CSS to style the font of 
the first letter / first line?  The examples show both - so are both 
acceptable?  We already know that <font> is deprecated and shouldn't be 
used, so you shouldn't be using that to style the first letter / first 
line.  Are these the only CSS pseudo-elements that should be mentioned?
2.2 Accessing alternative representations of content 
attr (alt):  There has been a lot of dialog on Bob Easton's Web site and 
on Joe Clark's book / Web site about how few people agree with the advice 
in this technique on the use of  attr(alt).  There needs to be a better 
example or explanation of why you would ever encourage anyone to use this.
:before and :after:  Again it seems like these are encouraged.  The 
drawbacks are not covered.  Agree they can be useful to provide additional 
information, but the techniques need to be clear that when used to provide 
important content, it may not be accessible.  First, if CSS is turned off 
you don't see it and second it's not supported in IE.
3.1 Media types .  Again, a lot of discussion on the Web that media styles 
are poorly supported and should not be used for certain media types.  I'm 
concerned about advancing this as a technique when it doesn't work as 
expected for many media types.
4.1 Creating borders .  Is this technique added to encourage people to use 
CSS borders instead of putting text in a layout table and specifying 
borders on the table?  Agree with editorial note that this doesn't map to 
any success criteria.
5.5 Creating Invisible labels for form elements . 
The User Agent notes should at least reference current versions of the AT. 
 The JAWS and WE versions are very old.
Agree with the idea mentioned in the Editorial comment to reference the 
title attribute technique in the HTML techniques page
With Firefox, the "hidden" labels using display:none are actually visible. 
 Given the inconsistency with display:none between AT and browsers, is 
this a good technique to recommend?
6. Generated content, automatic numbering, and lists .  Agree with the 
editorial note that there needs to be a cautionary note about misuse of 
CSS generated text.
7.2 Specifying color values by hex value or color name (optional) .  The 
User Agent information is very old and should be updated.  If current user 
agents support color, then is this technique really necessary?
7.4 Specifying background and foreground color .  How is this an 
accessibility issue?  It needs clarification.
8.1 Specifying fallback fonts.  This seems like a burden to ask authors to 
provide a fallback font.  If this is really just a JIS issue, can this 
technique be refined to include scenarios when fallback fonts must be 
provided?
9.4 Spacing letters and words .  The example looks like a usability rather 
than an accessibility issue.
9.5 Changing case   How is this an accessibility issue?  Needs 
clarification.
9.7 Underlining, overlining, and blinking .  No mention of underline and 
overline - are they okay to use? Believe they are except when used as the 
only way to convey meaning and that should be clear. If underline and 
overline are part of the technique title there should be at least some 
information about them in the technique. 



Becky Gibson
Web Accessibility Architect
                                                       
IBM Emerging Internet Technologies
5 Technology Park Drive
Westford, MA 01886
Voice: 978 399-6101; t/l 333-6101
Email: gibsonb@us.ibm.com
Received on Tuesday, 23 August 2005 14:07:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 17 July 2011 06:13:19 GMT