The ACT Rules Community Group (CG) writes testing rules for accessibility standards conformance. In the process, they find consensus in interpretation issues. This week, the CG published an “Implementations” page that shows these rules are not used in just one place, but in many: from manual methodologies to semi-automated (axe DevTools) and automated (axe-core, Alfa, Qualweb and more) tools.
Before going into why this is great, I will give some context. Accessibility standards like WCAG define criteria to conform with, they help makers of websites understand what a web without barriers means in practice. For instance, in WCAG, it means colours have sufficient contrast, pages have titles, buttons have names, and so on. In WAI-ARIA, it means elements with a
cell role must be contained in or owned by an element with the
row role, to name just one example. These, and all other requirements (“MUST”s) in standards, are testable requirements. This makes testing basically the process of finding out if you meet the standards (you, a product, a browser, a tool, an assistive technology, etc). It is a great lens to view accessibility through. and definitely a lens I use a lot when teaching.
But if you ever speak about the same issue with multiple accessibility specialists, you'll find that there can be differences in interpretation of these standards (see also: The Accessibility Interpretation Problem by Glenda Sims and Wilco Fiers). One way to avoid interpretation issues would probably be to have thousands of success criteria WCAG that allow for zero interpretation, but that imaginary world would be a much worse world. Instead, agreeing on how to test might be the best option we have.
The ACT Rules Community Group brings together makers of tools and other stakeholders to make testing accessibility standards more consistent, by finding agreement. They then write rules according to a format they specified. The rules that apply specifically to WCAG are also reviewed by AGWG, the group that produces WCAG.
This is tedious work and I, for one, am very pleased to see these rules now have implementations across different methodologies and tools. We'll probably see them more and more in the wild (I know I will in some tools I use). Congrats, Community Group, and keep up the great work!