Re: [tsvwg] Follow-up to your DSCP and ECN codepoint comments at tsvwg interim

[Lars Eggert <[e-mail]>]

Hi,

On 2020-3-11, at 17:50, Wesley Eddy <[e-mail]> wrote:
> There are 3 chairs, and I only speak for 1, but the bottom-line is simply "WG rough consensus" (decently described in RFC 7282).

I wonder (1) what the current status is and (2) if it can be summarized in the way to let me decide if I personally have an opinion.

I'm at least somewhat clueful about ECN and queuing, and I actually care about the topic (I want QUIC to work well in the presence of this, ideally taking advantage of it).

However, I have long ago given up on following the torrent of emails on this topic. There seem to be two camps that each believe they have a superior solution, fighting themselves to death over the details. More detail and more argumentation is unlikely to get me to a point where I feel comfortable on doing a hum.

I'd much rather see some short answers on some basic questions, such as:

Is the solution incrementally deployable?
Is it simple?
Does it offer better performance, to participating and ideally non-participating flows?
Is it at worst a no-op, or can it degrade performance for non-participating flows?
Can a partially deployed solution be rolled back?
Can it stay deployed but become a no-op if a different mechanism wins out in the end?
etc.

Because without some basic information along these lines, I fear I'd at the moment hum to not do anything, and that feels like we'd waste an opportunity.

Lars