Loading today’s stories
Product1 publisher3 min readPublished
Apple's WebKit team is taking Interop 2027 proposals now, and the entry rules filter harder than the final selection does, because a gap needs a stable standard and Web Platform Test coverage before anyone weighs how much it hurts.
The Product Desk · Product desk
Compiled by The Product DeskSomething wrong?How this is made
Most front-end repos carry a comment like "Safari needs this" or "Chrome only" above a line nobody wants to touch, and Apple's guidance now tells you which of those comments qualify for the proposal window [1]. Severity is not the gate. The gate is whether the behaviour is pinned to a stable standard and already covered by Web Platform Tests, because that shared suite is how progress on each focus area gets scored [5][8].
Most teams will fail that filter on their worst bug. Apple asks proposers to link the standard from the MDN page and to link existing WPT coverage, and where coverage is thin, to say how the tests could be written this autumn [11]. Anything short of sufficient tests or test infrastructure gets routed to Interop Investigations instead of a focus area [9].
The reward for clearing the bar is measurable. In Interop 2025 the selected tests went from 29 percent passing in every browser at the start of the year to 97 percent by the end [6], a gain of 68 percentage points [16] across the areas that were picked. Which is why being picked is the whole game, and the slots are not plentiful: 15 areas at the start in 2022, 20 in Interop 2026 [4][7], five more, roughly a third more capacity in four years [17]. The window itself runs 21 days counting both the open and the close [15]. What Apple wants inside those days is a specific feature rather than a theme, its own example being font-size-adjust instead of "typography" [10], plus the use case from your own work and evidence that the need is common, survey data included [12].
A two-by-two to run over your own bug list before you write anything. Axis one: can you link a stable standard. Axis two: can you link WPT coverage. Stable and tested is a focus area proposal, and the only work left is the case for how widely it bites. Stable but thin on tests means your useful contribution this month is tests, not prose [11]. No stable standard means Investigations [9]. Neither, and it is next year's fight, if the spec lands.
The published guidance sets out what a good proposal contains but does not say how competing proposals are weighed against each other or who makes the final call [18]. Community input has clearly landed before, since the 2026 set came partly from last year's open call [7]. The window functions more like a hearing that rewards the best-evidenced case than a ballot where votes are counted.
This is for the developer who already keeps the workaround documented and can point at the standard it works around. If your problem lives in a standard whose tests do not exist yet, the higher-value move this month is writing those tests, because that is the work the proposal form cannot do for you [11].
Ranked by verification strength, evidence, and original report placement.
Apple's WebKit team is inviting developers to submit their ideas for Interop 2027, the next edition of the cross-browser initiative aimed at improving web interoperability.
The Interop 2027 open call officially opened on September 3 and will accept proposals through September 23.
Apple has collaborated with Google, Microsoft, Mozilla, Bocoup and Igalia on Interop, an annual effort to make modern web technologies work more consistently across browsers.
Interop began in 2022, building on the earlier Compat initiative, with participating browser makers and organisations identifying 15 key areas where interoperability issues were causing problems for developers; Compat was conceived as a one-off, and Interop has returned every year since.
Each year the group selects a new set of focus areas, based in part on proposals from the web development community, and browser makers then work to improve their implementations in those areas, with progress measured through a shared set of Web Platform Tests.
In Interop 2025, the percentage of selected tests passing across all browsers rose from 29% at the beginning of the year to 97% by the end.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One outlet, one WebKit post, quoted at length
The dates, the 29%-to-97% pass rate and the five-point proposal checklist all come from a single 9to5Mac write-up of Apple's WebKit post, with the guidance quoted rather than paraphrased. That makes the facts easy to check against the original and impossible to corroborate here: no second outlet, and none of the other participating vendors speaks in our coverage.
Established program, unmeasured 2027 uptake
Interop has run every year since 2022 with Apple, Google, Microsoft, Mozilla, Bocoup and Igalia attached, and the 2026 edition carries 20 focus areas against the 15 the project started with, so the mechanism is plainly in use. The 2027 call itself has no numbers yet: our coverage gives no proposal volume from last year, no acceptance rate, and only 'partly' as the measure of how much of the 2026 slate came from the community.
Procedural reporting, one self-scored number
9to5Mac keeps close to Apple's wording and makes no claim the post does not, so the overshoot is small. The flattering figure is self-administered: 97% describes browsers passing tests the same group chose, and it appears without that qualification. Our own framing, that the entry rules screen harder than the final choice, holds only because the choosing is undocumented.
Participant's account, Apple-beat outlet
The underlying information is one participant describing an initiative it co-runs, relayed by an outlet whose beat is Apple and whose page closes with Amazon affiliate blocks. Neither pull is obviously bending the detail, since the guidance is quoted, but there is no counterparty present: nothing from Google, Mozilla or Igalia about what they want out of 2027, and no independent read on whether community proposals meaningfully move the slate.
Facts hold; the decision process is dark
Names, dates and quoted requirements from a public post are durable, and anyone can verify the deadline by opening the proposals tracker. The soft spot is interpretive rather than factual: how proposals are weighed against each other, and who settles the final focus areas, is stated nowhere in what we have.
product
Nvidia's server partners are passing the DRAM bill straight to Microsoft, Google and Oracle2 publishers
product
Amazon's $30 Echo Dot rise prices the door-opener, not the memory in it5 publishers
invest
A $2B bet that voice replaces the text box, not just the keyboard3 publishers
security
California hands app developers a four-bracket age signal starting 1 January 20271 publisher
Publishers with included, body-backed reporting in this cluster.
1 article · September 7, 2026