Re: [tsvwg] Dual-Q Polcing

[Sebastian Moeller <[e-mail]>]

Hi Gorry,

please see below.

> On May 27, 2020, at 16:41, Gorry Fairhurst <[e-mail]> wrote:
> 
> I'm going to leave others to pick-up on just a few details at the end.
> [...]
> 
> I would generally encourage people to separate security considerations from congestion control/traffic sharing issues.
> 
> Reducing claims about performance/features in drafts is always welcome. For the WGLC, we need to finally balance protocol design goals,

	[SM] Puzzled. L4S seems to offers no new protocol as part of the suite of RFC drafts. It does come with a few new requirements for protocols that want to get maximum ROI out of the L4S promises. Given that TCP Prague is not an RFC draft, what protocol design goals should we be concerned about here?


> against claims of what is offered,

	[SM] I am again puzzled, that very few members of this WG seem to have much issues with the L4S drafts promising the sky, while skimping on supplying robust evidence to support those claims.

> and doing this before WGLC is generally helpful.

	[SM] I would go further, it would have been helpful even before asking the ECT(1) as input or output question. Now, we are in the  weird situation, that L4S has basically gotten the "go" signal, with IMHO very little chance of not passing WGLC without any substantial changes to its core design. Verbiage, sure that might change a bit, but the core is by now pretty much accepted as set in stone (without sufficient evidence of it achieving its goals promises robustly and reliably).

> 
> The WG Chairs and ADs have been discussing what topics should best be included in which documents. My own suggestion would be to allow a few weeks for the document editors to digest this and then propose what needs to be done in the next steps to take this forward.

	[SM] Fair enough, IMHO there is no need to hurry any of this, or better no need to proceed at all before the requested robustness and reliability tests under to-be-expected conditions are performed, presented and evaluated, but I assume that that ship has sailed, and the process now will be based o the information and test results at hand?

Best Regards
	Sebastian


> 
> Gorry
> 
> G. Fairhurst, School of Engineering
>