performance.now().reject(reasons)

We’ve been at this make-performance-happen-now game for over a decade but the progress is slow. Optimizing images, critical path, etc. is a struggle; inconsistent adoption of HTTP/2 priorities signals industry hurdles we didn’t account for; we’re back at the drawing board with prefetch and prerender. What are the reasons, and what can we do to improve our odds as we head into the new decade? Let’s talk.