Home / Blog / Interop 2027 Voting Closes, Accessibility APIs Lead the Ballot
Accessibility
Interop 2027 Voting Closes, Accessibility APIs Lead the Ballot
The community ranking window for Interop 2027 shut on November 19, 2026. Among the hundreds of CSS, HTML and JS proposals under consideration, accessibility primitives like ariaNotify(), the focusgroup attribute and Shadow DOM referenceTarget drew some of the loudest public backing.
Interop is the yearly ritual that turns "already standardized but inconsistently shipped" into "actually works the same everywhere." Each cycle, browser vendors and the developer community nominate web-platform features with a stable spec but patchy cross-browser behavior, then rank which ones are worth the engineering effort to converge on. Submissions for Interop 2027 closed back in September, and the public ranking window that followed (Mozilla's ballot tool, plus a wave of personal rankings from working developers) closed on November 19, 2026. The official focus areas land in February 2027, but the ballots are already in, and teams that staff accessible, standards-compliant front ends by choosing to hire dedicated web developers who track this process get a head start on what's about to become reliably implementable.
What's on the ballot
Interop proposals need a stable specification and no outstanding objection from a browser vendor to qualify, which keeps the list to features that are already "real" on paper but inconsistent in practice. This year's ballot spans a wide surface: progressive-web-app installation, the sibling-index() and sibling-count() CSS functions, box-decoration-break unprefixing, JavaScript's Temporal API, and CSS layout additions like flex-wrap: balance, gap decorations and the random() function. But accessibility work showed up with unusual density on individual rankings, led by accessibility consultant Adrian Roselli's published ballot, which put three accessibility primitives at the top of his list.
Why accessibility proposals are leading
The three that recur across multiple public rankings share a pattern: each replaces a manual workaround that most teams get subtly wrong.
ariaNotify(): a scriptable way to push a screen-reader announcement without the current trick of mutating a hidden live region and hoping the assistive-tech stack notices the DOM change in time.
The focusgroup attribute: declarative arrow-key navigation for widgets like toolbars and menus, replacing hand-rolled tabindex and keydown-listener roving-focus logic that every team reimplements slightly differently.
Shadow DOM referenceTarget: lets an ARIA attribute like aria-labelledby point across a shadow boundary into a web component's internals, something that currently breaks silently for screen-reader and voice-control users the moment a design system adopts custom elements.
Roselli's framing for each is consistent: these aren't edge cases, they're places where "developers get it wrong" by default, because the correct behavior currently requires expertise most teams don't carry in-house rather than a markup attribute anyone can read.
What this means for developers
None of this ships from a ranking. Interop selection doesn't grant resourcing or guarantee an implementation; it's a prioritization signal, not a release. Still, it's a legitimate early read on where cross-browser accessibility behavior is likely to firm up over the next year:
if you're building design-system components on Shadow DOM today, track referenceTarget; it's the fix for the labelling-across-boundaries bug your team has probably already filed against itself;
toolbar, menu and tab-strip widgets built with manual roving tabindex are the first candidates to revisit once focusgroup lands, since it removes an entire class of keyboard-handling bugs rather than patching one;
don't wait on the spec to audit live-region announcements you already ship: ariaNotify() standardizes the mechanism, it doesn't retroactively fix an announcement that was wrong to begin with;
the final Interop 2027 focus list in February is the real signal for implementation timing; treat this ballot as "worth watching," not "safe to ship."
When does this ship / timeline
Proposal submissions for Interop 2027 ran from September 3 to September 23, 2026. The community ranking period that followed, including Mozilla's public ballot tool used by Roselli and dozens of other developers, closed on November 19, 2026. Per the project's own process documentation, the finalized scope of Interop 2027, meaning which proposals browser vendors commit to implementing and testing against, is scheduled to be published in the first half of February 2027. Nothing here is implemented yet; this is the ranking stage that feeds that decision.