Browsers do not share cache hits to CDNs anymore. The cache is sharded (not shared) across origins.

As of quite recently that may be changing in Chromium (early experiment): https://groups.google.com/a/chromium.org/g/blink-dev/c/9xWJK...

IIUC, the TLDR is that they're experimenting with loading a bunch of popular scripts (e.g. top `N` scripts from CDNs) on all browsers, so the response is immediate, but they can't use the timing for fingerprinting because it's immediate for everyone.

Of course it's a problem, that cache is not shared across domains, so it will be the initial experience of your users. Perhaps they can do some kind of lazy loading magic to hide it eventually

uncached, 8 seconds, cached, 6.5 seconds

Loading the entire wasm, touching all the things that need to be touched, etc, takes up the majority of the time

The good thing though is: because it's a WASM problem, not a Python problem, it will likely get better with time, and the benefits will automatically percolate to Python too.