Accessibility

Draft. The details in brackets must be filled in, and this text read by counsel, before launch. Until both are done it states our intent and binds nobody.

What we aim at

Level AA of the Web Content Accessibility Guidelines, version 2.2. This statement is voluntary — no law obliges a studio of our size to publish one — and it is written to be checked rather than believed.

What has been measured, and how

Every page of the site, in a real browser, at ten widths from 320 to 1920 pixels:

• No page scrolls sideways, at any of those widths.
• Every control that can be tapped is at least 24 by 24 pixels, the threshold of criterion 2.5.8.
• The whole site is operable from the keyboard: a skip link opens each page, focus stays visible, and the horizontal rows of projects and services are scrolled with the arrow keys as well as the mouse.
• The two carousels advance on their own. They stop the moment you point at them, the moment the keyboard enters them, and whenever the tab is in the background — and they do not run at all on a phone, or under the reduced-motion preference.
• The two scrolling bands and the two carousels stop under the reduced-motion preference — the bands on a full pass, so that they stay readable. Nothing flashes.
• One first-level heading per page, headings in order beneath it, and every form field tied to its label.

What falls short, and we know it

• Images carry no alternative text. Each one sits beside the words that name it — a project title, a person's name — and repeating them would say the same thing twice. Where an image carries something those words do not, that is a defect and we want to hear about it.
• The focus effect on the horizontal rows relies on a recent browser feature. Where it is missing, the row is simply uniform; nothing is lost but the emphasis.
• Three of the four legal pages, this one included, are drafts pending counsel.

What has not been done

No audit by a third party. No session with people who use assistive technology daily — which is the check that finds what automated measurement never will. No measurement on a real mobile device: the figures we have come from a machine, and they are optimistic.

Telling us

Write to [accessibility email], or use the form below, and say which page and what happened. We answer every message ourselves, and a barrier reported is a barrier we fix — this page is updated when we do.