Servo Baseline Readiness

16 hours ago

Servo is completing ~22 features/year, while the Baseline 'Widely Available' (BWA) set grows at ~52 features/year.
At current pace, Servo will plateau around 80% by ~2037 and never catch up to BWA.
Achieving full velocity parity requires ~44 FTE (€8.8M/year, €26.3M over 3 years).
A phased approach prioritizing high-usage features could reduce initial headcount and cost by up to 40%.
Key actions include unblocking 141 stalled features and fixing 51 regressions (features that lost >5 percentage points).
WPT pass rate improved from 30% to 62% over 2.5 years.
439 of 593 BWA features are measurable via WPT; 154 lack WPT mapping.
FTE cost is estimated at €200k/year, accounting for specialization and multi-disciplinary skills.
Usage-prioritized scenarios (e.g., deprioritizing <1% or <5% usage features) offer cost savings.
Data sources include wpt.fyi, web-features, and Servo git history.

Hasty Briefsbeta