Re: Update to WCAG 3

Hi Wilco and everyone,

As mentioned, we (the chairs) wanted to respond to your detailed feedback, and use it to create github issues and future discussions.

Some commentary below:

Conformance model

> The conformance model in this latest draft seems to be less certain than before even, with bronze / silver / gold levels now removed.

The conformance model in this draft is more certain in that we have eliminated many possibilities.  Much of the previous content was moved to the explainer.

Also, there are likely to be 2 types of conformance: Per-view (similar to WCAG 2), and a site-level conformance. The per-view model is straightforward, the site-level we can work on in parallel once more guidelines are fleshed out. I’ve opened an issue to track this:
https://github.com/w3c/wcag3/issues/154

Gaps & equity

> Already the first word of this draft fails this requirement: “W3C” is an acronym used before it is explained.

That’s a good point, although it might be that “W3C” becomes like KFC or NPR, where they are not an acronym anymore, they are the name. Whichever way that goes, we have committed to applying the guidelines to the document.

> If Unambiguous Text was part of WCAG 3.0’s foundation as proposed, it would make a big difference for equity towards cognitive… This is not likely to ever be accepted as “foundational” to W3C members, and so not a convincing demonstration of how WCAG 3.0 can be more equitable.

That doesn’t take into account one of the main changes in structure of WCAG 2>3, which is the supplemental guidelines with a (potential) scoring mechanism. A guideline may not be foundational, but if the bar set by regulators is above foundational, then there is more incentive for organisations to use those guidelines. It isn’t our place to tell regulators where to set the bar, but we should structure the guidelines so that bars of various levels can be applied.

> Next, Keyboard Focus Appearance doesn’t specify target size or contrast requirements.

This is something we’ve gone back-and-forth on, and I think the simplification of the decision trees in our last round may have gone too far. The documentation includes more details on testing focus indicators, but it’s too far down the structure at the moment.

That’s something we can work on, and is useful feedback for a future publication.
https://github.com/w3c/wcag3/issues/155

Rewrite existing requirements

> AGWG’s decision to attempt a full rewrite of WCAG 2.2 likely adding years to WCAG 3.0’s development. Proponents of this approach have long argued this is essential to achieve WCAG 3.0’s goals.

There needs to be a comparison to another choice here. I assume the alternative is to re-write WCAG 2.2?

Assuming the goals include being able to apply WCAG to a wider set of technologies, then I would argue it would take at least as long to re-write WCAG 2.2. As soon as you adjust basic assumptions such as web-page or the use of a user-agent, you would end up where we are: re-writing everything.

In order to move away from a ‘local maxima<https://www.interaction-design.org/literature/topics/local-maxima-in-ux>’, we need to adjust the basic assumptions. That means starting again, BUT we can and are using WCAG 2 to inform the text.

> Before the end of its charter, AGWG needs to demonstrate that the rewritten requirements are a worth-while improvement over WCAG 2.

That is not something we have committed to. That decision was made as a result of several long discussions about the future direction of WCAG, for improving coverage of both technologies and disability requirements.

Release plan

> The world needs to know whether WCAG 3.0 will be delivered in two, four or yet another eight years. AGWG can’t write a new charter without an agreed on timeline for WCAG 3.0

We decided to put 2 guidelines through the process to “mature” in order to have a data-informed schedule instead of guessing. It is unlikely we will have that until near the end of the charter, which is when we had committed to having it. We are on schedule to deliver this.

> Moreover, while AGWG has previously discussed options for breaking WCAG 3.0 up, or doing more incremental releases, so far no decision has been made public.

The most recent discussion (without resolution because we haven’t made a final decision) was on 3rd of  September<https://www.w3.org/2024/09/03-ag-minutes#t01>.  Until we have a viable alternative to publishing a single document as the base, we will continue with the current plan which is a single document. The final working group agreement will be when we have a proposed schedule.

Kind regards,

Alastair, Rachael and Chuck