On Wed, Oct 11, 2023 at 11:04 AM Yoav Weiss <yoavwe...@chromium.org> wrote:
> > So this will be sent in both H2 and H3? > Yes. As it stands right now, the Priority header will be sent in all cases where we prioritize (and already send frame-level priorities) on multiplexed connections (HTTP/2 and HTTP/3). The benefit of sending it on HTTP/2 is that it will let any endpoints that are not supporting the HTTP/2 dependency tree to start implementing support for at least initial priorities if they are interested in doing it (and for CDN's like Cloudflare that apply the extensible priorities-type scheme to HTTP/2 already to use explicit params instead of translating the browser-specific trees and weights from HTTP/2 into equivalent priorities). Eventually it would be worth experimenting with HTTP/2 priority update frames so sites that DO already support prioritization could switch completely over but that's out of scope for this particular change. > > > (which Chrome is the only browser to use for HTTP/3). > > > Do you know why other browsers don't implement the frame version? > The frame version is mostly in place to support reprioritizing in-flight requests (though it technically supports being used for an initial priority which is what Chrome does since Chrome also supports reprioritization). I'm not sure why other browsers decided to use the headers rather than the frames but there are a lot of possibilities where it makes sense (if the headers are easier to add, if the layering is such that QUIC is done in the OS-layer and priorities are done by the browser). > > Would be great to get an official word from them on what they are shipping > and when. Maybe file (non-blocking) positions on this? > Position for extensible priorities specifically or HTTP/3? Has Safari ever committed publicly to timelines? Position appears to be more than supportive since it's been implemented since Safari 14 or 15, the feature flag removed in 16.4 and it's mostly just questions around the rollout. There's some level of interaction with their IP-blinding proxy work that make it a bit more complicated. -- You received this message because you are subscribed to the Google Groups "blink-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAPq58w7Wsqnx9Pps2r9soUw1ACNKdCDk2C%3DU321w34iMvfsgdw%40mail.gmail.com.
