On Wed, Dec 21, 2016 at 5:52 PM, Martin Thomson <martin.thomson@gmail.com> wrote: > I think that what we need here is for someone to clearly articulate > what the model is and what is possible within that model. We don't > have that right now and I think we're struggling to reach conclusions > because we each have different expectations. > I would also add that from an HTTPbis perspective, HTTP/1 transport definitions are not likely to change. The vehicle for changing transport features is in subsequent revisions of the protocol (which might or might not imply QUIC) or perhaps extensions where possible. We made a pretty clean break in H2 between transport and general HTTP semantic things.. the latter typically have applicability across all versions of HTTP, but transport changes are not as well suited to looking backwards.
Received on Wednesday, 21 December 2016 23:01:06 UTC