Proposed changes to GL 3.1 L3 SC2 and L3 SC3

At the Team B meeting on September 14, 2005, we discussed updating GL 3.1 
L3 SC2 to include abbreviations and removing GL 3.1 L3 SC3 that is 
specific to acronyms and abbreviations.   Below is the proposal for the 
updated success criterion, the rationale and the open issues.   I'd like 
to see where we all stand on this SC change before tackling the changes to 
the guide doc and techniques. 

<proposal for updated GL 3.1 L3 SC2 (implies removing L3 SC3)> 
A mechanism is available for identifying specific definitions of words 
used in a restricted way, including idioms, jargon and abbreviations. 
</proposal> 

<rationale> 
Acronyms and abbreviations do not need to be called out separately from 
other definitions since they are examples of words used in a restricted 
way. 

The original L3 SC 3 seems like a very HTML specific success criterion 
since HTML has a specific mechanism for marking up abbreviations and 
acronyms. This seems to be a hold over from WCAG 1.0 and is no longer 
needed.   

The term acronym has been removed since an acronym is an abbreviation. 

The term unusual has been removed since it is not testable and is covered 
by "restricted way" (although there is disagreement on this, see the open 
issues section). 
</rationale> 

<open issues> 
Concern has been raised about removing the term "unusual" from the SC [1]. 
 John and I discussed this and still don't feel that "unusual"  is 
testable - what are the other opinions about removing this phrase?  John 
and I believe that restricted covers the range that we need since 
restricted can be defined in the following manner (my paraphrase of email 
conversations): 
What is meant by using a word "in a restricted way"-- the author has to 
"restrict"/limit meaning to a particular definition rather than leaving it 
open. For example, in the US the acronym "ADA" can stand for at least 
three different things: the Americans with Disabilities Act, the American 
Diabetes Association, and the American Dental Association. And in lower 
case it's the name of a programming language, but it's *not* an acronym-- 
the language is named for Lady Ada Lovelace, a 19th-century English woman 
who did some pioneering work in computation! 

The word is used in a way that differs from the primary dictionary 
definition 

The word is used in the context of a specific subject matter.   

Since it does not specify,  does this SC imply that all occurrences of the 
definition should be programmatically determined?  For abbreviations we 
currently specify the frequency in the technique by stating that only the 
first occurrence needs to be marked (there is also disagreement here).   
</open issues> 

<proposed general techniques> 
Using inline expansion 
Using a Glossary page 
Searching an online dictionary 
Using a Dictionary Cascade 
</proposed general techniques> 

<proposed HTML techniques> 
Link to a glossary 
Use <abbr> element 
Use <acronym> element (propose marking this as deprecated in favor of just 
<abbr>) 
Supplemental meaning clues 
Using a definition list 

One of the issues of combining the two success criteria is that the HTML 
techniques are dependent upon the type of content. For example, the HTML 
techniques for using the <abbr> or <acronym> element are specific to those 
types of words and will not work for jargon.  Likewise, definition lists 
are not the best choice for abbreviation expansion.   John and I think 
this is acceptable since the author should know what type of content is 
being generated and use the appropriate technique. 
</proposed HTML techniques> 

thoughts? 

[1] 
http://lists.w3.org/Archives/Public/public-wcag-teamb/2005Sep/0066.html 

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 Thursday, 15 September 2005 18:03:08 UTC