Will HTTP-WG keep messing around with the draft until our ClientHello header is filled up with the ALPN extension? h2-14 was left because it was an implementation draft. But are h2-15 and h2-16 need to be advertised?
Summary: Advertise h2-16 in parallel with h2-14 and h2 → Advertise h2-17 in parallel with h2-14 and h2
I expect -17 to go to the rfc editor, which means it will really be final as h2 and no-one will try and negotiate -17 in particular instead of -17. So we can probably wontfix this bug, but I'll wait for the standards action to happen before closing this. We can kill all the draft versions soonish. gecko 40 sounds about right. fwiw they've all been used at various points with overlap at the same point between all 3 versions. "implementation draft" isn't really a thing in practice and it really doesn't make a lot of sense if you want to get real feedback on the changes.
based on -17 going to proposed standard, we can change this bug to be remove draft-standard support starting in gecko 40. i.e. h2 only. (this should not impact spdy - that's a different bug and a different timeline.)
Summary: Advertise h2-17 in parallel with h2-14 and h2 → remove h2-draft support starting in gecko-40
I've been looking forward to this ever since the day I started adding all those alpn tokens :) I'll write the patch, and we can have it hanging around waiting for 40 to be on m-c.
Assignee: nobody → hurley
Created attachment 8573426 [details] [diff] [review] patch
Attachment #8573426 - Flags: review?(mcmanus)
Huh, git-bz fail - it put my comment into the commit message of the patch instead of the comment (maybe I put it in the wrong place in the file, who knows). Anyway: This should hit all the high points. Obviously, I'll wait until m-c becomes gecko 40 to land (assuming this is r+'d) https://treeherder.mozilla.org/#/jobs?repo=try&revision=e119e99603ef
Comment on attachment 8573426 [details] [diff] [review] patch Review of attachment 8573426 [details] [diff] [review]: ----------------------------------------------------------------- landing is contingent on having a RFC number to point to
Attachment #8573426 - Flags: review?(mcmanus) → review+
Added RFC numbers in appropriate files. Hooray! remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/dc3c55cedb3f
Status: NEW → RESOLVED
Last Resolved: 3 years ago
status-firefox41: --- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla41
Added the site compatibility doc: https://www.fxsitecompat.com/en-US/docs/2015/http-2-draft-is-no-longer-supported/
Keywords: dev-doc-needed, site-compat
Keywords: dev-doc-needed → dev-doc-complete
You need to log in before you can comment on or make changes to this bug.