bekee gibson

Accessibility

Accessibility on this site.

I take accessibility seriously in my work, which means I am also responsible for it on my own site. This page is the honest version: what I test, what I do not, and what to do if you find a problem.

There is a lot of shorthand in the sections below. If a term is new to you, it is defined in plain language in the accessibility glossary.

What I test.

Every page on this site is scanned with a multi-lane audit on each significant change. Open issues are tracked and fixed. Findings from the automated and AI-assisted passes are reviewed by a human (me) before any change ships.

The audit covers, at WCAG 2.2 Level A and AA:

  • Image alt text, both presence and quality (SC 1.1.1)
  • Programmatic structure, landmarks, and ARIA correctness (SC 1.3.1, 2.4.1, 4.1.2)
  • Form field labels and autocomplete (SC 1.3.5, 3.3.2)
  • Color contrast for text (SC 1.4.3)
  • Text-spacing tolerance (SC 1.4.12)
  • Use of color and graphics contrast in forms, charts, and non-decorative graphics, surfaced for human review (SC 1.4.1, 1.4.11)
  • Keyboard support where it can be automated (SC 2.1.1)
  • Page titles, both presence and descriptiveness across the site (SC 2.4.2)
  • Link purpose in context, both presence and meaningfulness (SC 2.4.4)
  • Multiple ways to find a page across the site (SC 2.4.5)
  • Headings and labels, both empty-heading detection and descriptiveness (SC 2.4.6)
  • Page language declaration (SC 3.1.1)
  • Consistent identification of the same destination across pages (SC 3.2.4)
  • Accessible-name quality for non-link controls (SC 4.1.2)
  • Whether the skip link actually works: its target exists, it reveals itself on focus, and it comes first in the tab order (SC 2.4.1)
  • Focus order, walked by actually pressing Tab and recording where focus lands (SC 2.4.3)
  • Focus visibility, confirmed by comparing the rendered pixels focused against unfocused, never by reading the stylesheet (SC 2.4.7)
  • Meaningful sequence, surfaced where the observed focus order disagrees with the visual order (SC 1.3.2)
  • Pages carrying animation, surfaced for human review; the tool never decides whether motion can be paused, and it never measures flash rate (SC 2.2.2, 2.3.1)
  • Touch and pointer target size, both the minimum and the spacing that can stand in for it (SC 2.5.8)
  • Whether a focused element is buried under something else, tested by asking the page what is actually painted on top rather than by reading the stylesheet (SC 2.4.11)

Beyond the WCAG 2.2 AA success criteria, the automated scan also runs axe-core’s best-practice rules. These catch problems that are not strict WCAG failures but still get in a real person’s way: a heading order that skips levels, a page with no landmark regions for screen-reader navigation, and similar gaps. I report these separately and label them “best practice,” not as a success criterion, so you can always tell a conformance requirement from a recommendation.

What I do not test.

  • I do not run real assistive-technology user testing. I do not have screen-reader users, voice-control users, or switch users on call. That kind of testing is the only way to confirm a real person can use the site. If you use assistive tech and run into something, please tell me (see below).
  • I do not test the site across every browser; the automated scan runs in Chromium via Playwright. The scan is not a substitute for testing in actual assistive-technology software.
  • These specific WCAG criteria are out of scope of my automated coverage: Reflow (SC 1.4.10) and Content on Hover or Focus (SC 1.4.13). Viewport-resize behavior and hover and focus popups need a human.
  • The keyboard walk sees the page as it loads. Closed menus, modals, and accordions are invisible to it unless the interaction is scripted, and that is the largest gap in the lane.
  • Reaching a control is not the same as operating it. Whether Enter activates it, whether Escape closes it, and whether focus returns somewhere sensible afterward are all still manual.
  • Keyboard traps (SC 2.1.2) are not tested. A walk would stall inside a trap rather than identify one, so a trap shows up as a truncated run rather than as a finding.
  • WCAG 2.2 added nine success criteria. I cover two: target size (SC 2.5.8) and focus not obscured (SC 2.4.11). Four are out of scope for now, because each needs an interaction or a multi-step flow a single page load cannot show: dragging movements (SC 2.5.7), consistent help (SC 3.2.6), redundant entry (SC 3.3.7), and accessible authentication (SC 3.3.8). The remaining three are Level AAA.
  • The focus-obscured check presses Tab to the end of the page and then Shift+Tab back up, and a finding says which direction it turned up in. Tabbing forward scrolls a target toward the bottom of the screen, well clear of a header pinned at the top, so a sticky header tends to cause trouble on the way back up rather than on the way down. The backward walk stops if focus moves into an embedded widget it cannot follow, like a captcha in an iframe, and the report records where it stopped rather than calling the rest clean.
  • I do not claim 100% accessibility. Nobody can.

The standard I aim for.

WCAG 2.1 Level AA. A handful of Level AAA items are met where it is reasonable to do so; I do not claim full AAA conformance.

Third-party content.

This site embeds Cloudflare Turnstile, the anti-spam check on the contact form and on the two Site Health Scan forms. That is the captcha named above: the backward keyboard walk cannot follow focus into its frame, so on those three pages the walk stops there and the report says where it stopped. Automated checks will sometimes flag issues inside a third party’s markup that I cannot reach, and if I add another embed, a video player or a calendar, the same applies. I disclose it on the affected page rather than pretend.

How to report a barrier.

The quickest way is to say hello. Please include the page URL, what you were trying to do, what assistive technology you were using (if any), and what happened. I usually reply within a couple of days. If something is blocking you completely, say so at the top of your note; I treat that as a priority.

Last reviewed.

2026-08-16