what we’ve tested, and what we haven’t.

the claim, stated honestly

We aim at WCAG 2.2 level AA. Today we claim partial conformance — not because we know of a failure, but because our testing has been automated plus hands-on measurement, and nobody has yet driven this site with a real screen reader or tested it with disabled people. Until that happens, “we found no failures” and “there are none” are different sentences, and we are only entitled to the first.

what has actually been measured

All eleven pages of this site, in a real browser, not asserted from a checklist:

• axe-core: zero violations on every page, at 1440px, 390px and 320px, against the WCAG 2.0 and 2.1 A and AA rule sets.
• Contrast: 17 distinct text-and-background pairs, none below its floor. Each one computed from what the browser actually renders, using the WCAG relative-luminance formula. The lowest is 5.26:1 on large text (floor 3:1); the lowest on normal-size text is 5.76:1 (floor 4.5:1). Most of the site runs at 16–18:1.
• Reflow (1.4.10): passes. No horizontal scrolling at 320 CSS pixels, which is the same thing as 400% zoom on a 1280px window. Checked at fourteen widths from 320 to 1920px, and at seven landscape phone sizes.
• Text spacing (1.4.12): passes. With line height forced to 1.5, letter spacing to 0.12em, word spacing to 0.16em and paragraph spacing to 2em, no page overflows and no text is cut off.
• Target size (2.2 AA, 2.5.8): passes. Every non-inline control is at least 24×24 CSS pixels, and most are 44×44. The navigation chips are drawn tight on purpose, so each one carries an invisible box extending its tap target to 44px — checked by probing the real hit area 21px above and below each chip’s centre, on every page, rather than by measuring the visible ink. Where the nav wraps to two rows on the narrowest phones, the rows are spaced so that no two of those targets overlap.
• Keyboard: a skip link, a visible focus ring on every control on both colour schemes, and the crisis panel traps focus while open, closes on Escape, and returns focus to whatever opened it.
• Reduced motion: with prefers-reduced-motion set, the drifting names behind the homepage headline are removed from the page entirely rather than paused, and nothing else on the site animates.
• Increased contrast: with prefers-contrast: more, secondary text and rules step up to higher-contrast values.
• No third-party requests and no cookies when a page loads — verified per page. Fonts are self-hosted. There is no analytics, no tracking and no consent banner to fight with. The only outbound request the site ever makes is when you submit the email form.
• Text sizing: every font size on the site comes from a scale bounded in rem, never raw viewport units, so browser text-size settings and zoom both work.

what we have not tested

This is the part most accessibility statements leave out.

• Real screen readers. Nobody has run this site through NVDA, JAWS, VoiceOver or TalkBack. Automated tools check that the markup is well-formed; they cannot tell you whether the result makes sense read aloud.
• Voice control (Dragon, Voice Control), switch access, and screen magnifiers at high magnification.
• Real people. No disabled person has tested this site. That is the gap that matters most, and no amount of automated testing closes it.
• Cognitive load and reading level. The evidence map in particular is long, dense and full of statistics. It is almost certainly harder to read than it needs to be.

known rough edges

• Four of the wide data tables on the evidence map scroll sideways on a phone, because their content is genuinely wider than a phone. Each one is a labelled, focusable region so it can be reached and scrolled from the keyboard, and a hint appears only when a table really does overflow — but sideways scrolling is still worse than not needing to.
• Headings use a newer CSS trimming feature that only some browsers support. On the others they simply sit in a slightly looser box. Nothing is lost.
• The drifting names behind the homepage headline are decorative, hidden from assistive technology and removed under reduced motion — but if they are distracting for you and reduced motion is not set, that is a real problem and we would like to know.

where we stand legally

The Public Sector Bodies (Websites and Mobile Applications) Accessibility Regulations 2018 apply to public sector bodies. We are not one, so they do not bind us, and we are following them anyway because they are the right standard.

The Equality Act 2010 does apply to us. It requires reasonable adjustments so that a disabled person is not put at a substantial disadvantage. If you think we have failed that, tell us — and if we do not put it right, the Equality Advisory and Support Service can advise you, at equalityadvisoryservice.com.

tell us something is broken

Email hello@isjamesok.org. You do not need to know the technical term for it, and “this bit is unreadable” is a complete and useful bug report. Tell us what you were using if you can — the browser, the phone, the screen reader — because that is usually the hard part to guess.

We are a very small team, so we cannot promise a fix by a date. We can promise that we will not argue with you about whether it is a problem.

If you need something on this site in another format — larger print, plain text, or read aloud — ask, and we will send it.

what happens next

Before launch: a real screen-reader pass, and testing with disabled users. Those two are the difference between the partial claim above and a full one, and neither has a date yet. When they are done, this page says so and says what they found — including anything that fails.

Safeguarding · Privacy · ← Back to isjamesok?