Accessibility statement
This is an English translation, provided for readers of the English documentation. The Dutch version is the binding one: it is the legally required form for a Dutch government site, and where the two differ, the Dutch text prevails.
The statement describes how far RegelRecht meets the accessibility requirements
for government websites. The legal standard right now is WCAG 2.1 level AA, via
EN 301 549 and mandatory in the Netherlands under the Besluit digitale
toegankelijkheid overheid. WCAG 2.2 has been published by the W3C but is not yet
part of EN 301 549; once the European standard moves to 2.2, the Netherlands
follows. RegelRecht has therefore already tested against the nine new 2.2
criteria, ahead of that transition. The statement covers the site served at
regelrecht.rijks.app and at docs.regelrecht.rijks.app, so the landing page,
the sign-up form, and the documentation.
This is a draft. The status below rests on an automated test in the build pipeline and a manual review by the team. No independent party has audited the site, and the statement is not yet listed in the DigiToegankelijk register. Until both of those are done, this is not a legally valid statement and RegelRecht carries no accessibility label.
Compliance status
Partial. The automated test finds no errors on the criteria it measures, and the manual review covers the criteria no tool tests reliably. A few known limitations remain, listed below. Without an external audit, RegelRecht claims no full compliance.
Language of the site
The documentation is mostly English; the landing page and the sign-up form are
Dutch. Every page carries a lang attribute that matches the language of its
content (en for the docs, nl for the landing and the Dutch statement), so a
screen reader picks the right pronunciation. Where a single term in the other
language appears within a page, that fragment gets its own lang. This covers
WCAG 3.1.1 (language of page) and 3.1.2 (language of parts).
How this was tested
Automated, on every change. An accessibility test runs in CI on every change
that touches the documentation. It builds the site and runs
pa11y-ci with two independent engines,
HTML_CodeSniffer and axe-core 4.11, against every generated page. The URL list
comes from the build, so a new page is tested automatically and does not slip
past the check. The test runs locally with just docs-a11y.
Those two engines cover a large part of what can be measured by machine: contrast, form labels, heading structure, landmarks, and alt text. What a tool does not see reliably was checked by hand.
Manual, including a head start on 2.2. The team went through the nine success criteria that WCAG 2.2 adds, although they are not yet a legal requirement: focus not obscured (minimum and enhanced), focus appearance, dragging movements, target size, consistent help, redundant entry, and accessible authentication (minimum and enhanced). Beyond that, hand-tested:
- keyboard-only operation, including the skip link and the focus order;
- dark mode, for contrast and legibility;
- rendering at 200% zoom and at 400% reflow.
Known limitations
The automated contrast test flags a few elements whose contrast actually passes comfortably. The engines cannot read the ratio of these elements reliably. The team measured the real ratio in the browser and excluded these elements from the automated check, with a note in the configuration:
- Diagrams (Mermaid). Diagrams render as inline SVG, with text on transparent layers where axe cannot read the colour behind it. The measured ratio is 14.4:1 for text in flow and state diagrams (dark blue on light blue) and well above the requirement for the C4 diagrams (white on dark blue). The diagram colours come from the NLDD palette.
- Code samples. Code blocks carry a light and a dark theme on the same element; the test mixes the two colour sets and measures a blend that never appears on screen. The text that does appear clears the 4.5:1 that AA asks for in both modes.
- The hero on the landing page. The title text is white on a dark-blue gradient. axe cannot judge a gradient. The weakest point of the gradient, its lightest stop, gives white-on-dark-blue a ratio of 11.4:1; towards the dark end it rises to 15.5:1.
The accessible name of each diagram (role="img" with an aria-label) is checked
separately against the build output in the same gate, apart from the excluded
contrast measurement.
Not yet covered
- There has been no independent audit. Everything above rests on the team’s own test and review.
- The interface uses web components from the NLDD design system that render their content in a shadow DOM. Their own focus indication and internal accessibility fall outside the site’s control and depend on the design system.
Reporting a problem
Run into an accessibility problem, or something does not work? Email regelrecht@minbzk.nl. Describe what went wrong and on which page, and we will pick it up.
Drawn up
This draft statement was drawn up on 21 May 2026, based on the test and review of that moment. RegelRecht is an exploration and still under development; the statement is updated when the site changes or after a formal audit.