numbering proposal

Hello,

On the mailing list, we have had some discussion about numbering success 
criteria [1].  Jason, Gregg, Ben, and I were editing the next draft and 
came up with the following proposal.

If each guideline was identified by a letter (e.g., "N" for "Navigable") 
rather than a number (3) and if each conformance letter was identified by a 
letter (M for Minimum, S for Second Level, T for Third Level), then each 
success criterion could have a unique identifier (e.g., N3M1, N3M2, N3S1).

For example, the current checkpoint 3.4 with the proposed scheme:
Checkpoint N4 Provide consistent and predictable responses to user actions.

Success criteria

You will have successfully met Checkpoint N4 at the Minimum Level if:
   N4M1. where inconsistent or unpredictable responses are essential
        to the function of the content (e.g. mystery games, adventure
        games, tests, etc.) the user is warned in advance of encountering them.
   N4M2. wherever there are extreme changes in context, one of the following
       is true:
       a. an easy to find setting, that persists for the site visit, is 
provided
           for the user to deactivate processes or features that cause
           extreme changes in context or
       b. extreme changes in context are identified before they occur
           so the user can determine if they wish to proceed or so they
           can be prepared for the change

At the Second Level:

   N4S1. the content has been reviewed, and it has been found that where
       inconsistent or unpredictable responses are essential to its
       function (e.g. mystery games, adventure games, tests, etc.),
       the user is warned in advance of encountering them

At the Third Level:

(presently no additional criteria for this level.)
===

Pros and Cons:
   - Should sound better for screen reader users since there are not
      multiple numbers that are read in a row and the numbers follow
      the letters.
   - Creates a unique identifier for each success criterion.
   - Might be harder to decipher the checkpoints.
   - Is not consistent with WCAG 1.0, ATAG 1.0, or UAAG 1.0
     numbering schemes.
   - However, since it is different and since some people need to
     conform to both WCAG and ATAG or WCAG and UAAG,
     they are less likely to confuse ATAG/UAAG checkpoint 3.4
     with WCAG 2.0 checkpoint N4.
   - Not sure that it help distinguish our conformance levels from
      those of WCAG 1.0 (Level A, Level AA, Level AAA) and
      the priority scheme (priority 1, 2, 3).  Although, we did make
      a minor change: previous "Level 2" is "Second Level"
      and "Level 3" is "Third Level."

Discussion questions:
   1. Is this easier or harder to decipher than the previous proposals?
       i.e., 1.1.1.1 or 1.1-1a ?
   2. Is it a benefit that this is different from the WCAG/ATAG/UAAG 1.0
       numbering scheme or will it cause confusion?
   3. Our preliminary tests showed this was better for screen
        reader users.  Do you agree or disagree?
   4. Other issues or comments about this proposal?

We discussed some other possibilities but discarded those because:
1. we wanted to associate the checkpoint with the guideline in some way to 
preserve the grouping (thus we didn't want to assign a unique number to 
each checkpoint).
2. we wanted to make something that would not be garbled by a screen reader 
(1.1.-1a was read as "one one one ah" instead of "one one one A").


[1] http://lists.w3.org/Archives/Public/w3c-wai-gl/2002OctDec/0266.html
[2] http://www.w3.org/WAI/GL/WCAG20/#navigation
[3] http://www.w3.org/WAI/GL/WCAG20/#consistent-responses

-- 
wendy a chisholm
world wide web consortium
web accessibility initiative
http://www.w3.org/WAI/
/--

Received on Friday, 14 March 2003 16:48:12 UTC