Closed LPardue closed 4 years ago
Hello both,
Sorry for the late reply, have something else I need to finish today.
This PR looks good overall, I would just have included a section indicating that research and experiments have shown that simpler schemes can achieve similar performance as the more complex, dependency-tree driven schemes we've seen in the wild for HTTP/2. And that, as such, this initial document focuses on a simple and straightforward scheme as a result. If you agree, I can make a new PR for that, since this one was just merged :)
@rmarx That sounds like something highly desirable to clarify. I think that a PR would be more than welcome.
On the priority team mailing list I advocated for a consolidation of the problem statement
This PR builds on top of #58, creating a new section that combines that text, with text from the introduction and pulls in some of Roberto's text. All of that happens on https://github.com/kazuho/draft-kazuho-httpbis-priority/commit/b2b927ccbeea76f9edcb5e49ad9be06530d9e2ab
@rmarx I didn't pull anything specifically from your paper but if you feel there is something that I don't cover then please do suggest it