[tsvwg] Results of consensus call on ECT(1) usage

[Wesley Eddy <[e-mail]>]

**

*Thank you to all of the participants that provided feedback to our 
consensus call on ECT(1) usage.  After discussing the contributions 
between the chairs and our ADs, here are the conclusions we’ve drawn.*

*

Result of consensus call: TSVWG has interest in working on a low-latency 
service using ECT(1) as an input (e.g. L4S) in preference to working on 
enabling high-fidelity congestion control using ECT(1) as an network 
output signal (e.g. SCE). The chairs see strong support for this 
direction, including significant interest from network operators and 
vendors.

Specific concerns that still need to be worked on (not an exhaustive list):

 1.

    The WG needs to consider operational issues in L4S deployment and
    provide guidance to operators of networks that participate in the
    L4S experiment and operators of networks that do not.

 2.

    The WG must ensure that L4S does not add significant new incentives
    for network users or operators to lie or cheat in packet markings,
    beyond marking incentives for current ECN functionality.


Path forward:

  *

    The objective is to advance the L4S drafts to WGLC after revising
    them to resolve technical issues.

  *

    A successful WGLC on the drafts will require TSVWG rough consensus
    that the L4S experiment is safe to perform on the Internet.

  *

    Based on the discussion to date, the chairs recommend that a draft
    describing TCP Prague be produced, to provide an example of endpoint
    congestion control that is suitable for use with L4S.

*