This page states the site’s accessibility target, how progress towards it is measured and where the known limits are. Nothing unmeasured is called compliant; what has been measured is stated together with the method. If you meet a barrier, the reporting route at the end of the page is open.
Target and testing
The target is level AA of the Web Content Accessibility Guidelines (WCAG) version 2.1. That is a testing target rather than a declaration: before each release the pages are walked through from start to finish with the keyboard alone, colour contrast is measured with an automated contrast tool, and the heading hierarchy and landmarks (banner, main, navigation, contentinfo) are checked. The results are applied to the page template; when the template changes, the tests are repeated. Testing with real screen reader users has not yet been done; that is listed among the known limits below.
Keyboard navigation
Every page starts with a “skip to main content” link; it becomes visible on the first press of the Tab key and leads to the main content. The focus ring is visible on every link, button and form field; focus is never hidden. Wide tables are placed in a horizontally scrollable region so that they do not overflow at 360 pixels, and that region can be scrolled with the keyboard as well. Collapsible sections use the browser’s own disclosure element; when a link leads into a closed section, the section opens automatically. Transition animations are switched off for users who have set a reduced-motion preference.
Typeface, contrast and readability
Headings are set in Literata; the typeface is served from the site’s own server and, if it fails to load, the browser’s serif typeface takes over, so text is never invisible. Text can be enlarged with the browser setting; at two hundred per cent the page needs no horizontal scrolling. Contrast between text and background is kept above the AA threshold and measured before each release. Information is never conveyed by colour alone: the status shown in boxes such as the country status box is written in words next to the colour. The language of every page is marked, and the links between language versions carry the language of the target page, which matters to readers switching between Turkish, English and French.
The checkers without JavaScript
The checkers on this site run in the browser with JavaScript. If JavaScript is off or fails to load, an explanatory text appears in the checker’s place saying what the checker tests and what it does not; the rest of the page and the body text read in full without JavaScript. The checker itself can be used with the keyboard: the select boxes are labelled, the result section carries a heading, and the result is added to the page in a way that screen readers announce. The quote form can also be submitted without JavaScript.
Known limits
Some things have been measured and found wanting; others have not yet been measured. Testing with real screen reader users has not been done. Some external links on the site lead to legislation in PDF form and to official pages; the accessibility of those documents is outside this site’s control, and some of them may be scanned images. Wide comparison tables need scrolling on small screens; that was judged less bad than breaking the table apart, but it is not ideal. On the English and French pages, Turkish institution names are left in their original form; a screen reader will pronounce them according to the page language.
How to report a barrier
If a page cannot be used with the keyboard, a piece of content cannot be understood with a screen reader, or a contrast is too low, write to the email address on the contact details page; the page address, the assistive technology you use and the barrier you met are enough. The report is measured, and if it holds, the template is corrected and the correction applies to every page. On request the content of a page is sent as plain text by email; that is a temporary route until the correction is made, not a permanent solution.