Idea for additional Level A criteria

I don't disagree with any of the discussion points below.  I do want to offer the idea that WCAG 2.3 could add 5 or 6 single A SC, and they might not even be all that hard to write!

IHMO there is a case to made that the biggest defect with 2.0 is that 1.3.1 Info and Relationships is (1) too much of "kitchen sink", and (2) the requirements for meeting 1.3.1 are not as explicit as perhaps they could be.

I suggest we consider adding SC under 1.3.1.

Just for example, to illustrate what I think we could discuss:


  *   1.3.1.1 Data in Rows:  The header for tabular data row is programmatically determinable.
  *   1.3.1.2 Data in columns:  The header for tabular data in a column is programmatically determinable.
  *   1.3.1.3 Complex Tables:  For tabular data table with more than one header, headers are programmatically determinable.
  *   1.3.1.4 Hierarchical Relationships:  Hierarchical (parent-child) relationships are programmatically determinable.

I started with these four from https://www.access-board.gov/ict/#502.3.1 and there is more (e.g., forms) to mined from Understanding<https://www.w3.org/WAI/WCAG21/Understanding/info-and-relationships.html>.


From: Alastair Campbell <acampbell@nomensa.com>
Sent: Tuesday, April 27, 2021 2:55 PM
To: WCAG <w3c-wai-gl@w3.org>
Subject: RE: WCAG 2.3 v 3.0 - was RE: XR Subgroup Update [April 27th 2021]

Hi everyone,

(Chair hat off), two points I find compelling from JonA:

> without personalization or a non-binary model these great recommendations likely will not get consensus in WCAG 2.x and will take up a lot of time ...  Is getting 5 to 6 new A or AA criteria in WCAG 2.3 of value over pushing WCAG 3.0 additional years down the road?

To add to that point:
The 2.x structure is set, and it is getting harder and harder to get new SCs in.
In WCAG 2.1 we started with over 60 SCs, and 17 were incorporated.
In 2.2 we are looking at 9 (1 of which is a AAA variation).

On that trajectory, a 2.3 would likely have 4 new criteria unless we found new requirements that were easy to translate to criteria. (In which case, where have they been?)

Tested & scoring is more flexible in WCAG 3.0 so it could be quicker to get new requirements into 3.0 than 2.3, even with the longer timeline assumed for 3.0.

The new requirements included in 3.0 could also be of more use to people with disabilities if they can use different testing methods, and/or better account for task and context. That's an important quality metric I think we sometimes gloss over.

Even for a fresh requirement that we haven't considered before, creating a WCAG 3.0 guideline and WCAG 2.3 criterion is double the work of just doing one. Putting them into different structures requires you to create it separately for each, even if you start with a good requirement and appropriate ACT tests.


>  if [WCAG 2.3] exists then that pulls their expertise away from 3.0 meaning 2 years from now they have to be brought back in to the 3.0 cycle when their expertise was needed earlier.

Indeed, now that we are at the point where parallel working can happen on 3.0, time spent on 2.3 would likely add to the 3.0 timeline. That does depend on what members want to work on, but it is a factor.

Of the ~27 people on the call today I think 3 raised their hand for working on 2.3 criteria. If anyone else is interested in that please do us know at group-ag-chairs@w3.org<mailto:group-ag-chairs@w3.org>

Kind regards,

-Alastair