Re: [quicwg/base-drafts] Consider a new stream type to act like MAX_PUSH_ID (#2418)

[Jana Iyengar <[e-mail]>]

This is a nice idea.

Encoding deltas avoids the server having to check if the new MAX_PUSH_ID is
smaller, but meh.

The only issue I see here is HoL blocking. It would be nicer to not
introduce dependencies and send these updates in DATAGRAM frames, but of
course, we don't have those.

On Tue, Feb 5, 2019 at 5:01 PM Lucas Pardue <[e-mail]>
wrote:

> If the sequence is monotonic increasing, a series of deltas will make it
> more efficient. Slightly. I mean, the stream overhead here will dominate
> anyway.
>
> Neat idea. unsigned deltas may actually be a more natural mapping for what
> the client is trying to express in terms of managing concurrent pushes.
> However, this might add some work for the server to reconstruct the push
> ID. Not sending explicit values seems like an easy way to misunderstand the
> clients intent.
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <https://github.com/quicwg/base-drafts/issues/2418#issuecomment-460604870>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AKjg1I6DUGrP6xli1nutis540wJZ4rTzks5vKWudgaJpZM4ai5xW>
> .
>


-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2418#issuecomment-460609051