Re: [tsvwg] plan for L4S issue #29

[Gorry Fairhurst <[e-mail]>]

See below.

On 29/09/2020 16:49, Rodney W. Grimes wrote:
>> On Mon, 28 Sep 2020, Gorry (erg) wrote:
>>
>>> At some point the working group needs to publish the spec. - This final
>>> stage is taking longer than I would hope, and I do hope that will be
>>> seeing a WGLC soon.
>> Do we actually?
Yes.
>> I still haven't ruled out that we decide not to use these bits, for now,
>> because we don't know enough how it will affect the entire Internet.

Still possible, if the WG as a whole decides that.

> I can support the "we don't know enough" statement.
>
> Sadly this was not presented as a choice during the "vote" on
> ECT(1) input vs output, but it should have been.  In fact I thought it
> was going to be.  Alisha Cooper clearly mentioned it during her suggestion
> to NOT attempt to take a consensus call during the end of the virtual
> meeting due to lack of time.  Yet what came out was a different
> 3rd option, which personally I just do not understand as it was
> extremely narrow.  Further what was presented to SCE as a selection
> between ECT(1) as an input or output without any concern over this
> being L4S vs SCE was clearly NOT presented to the WG in that light,
> but instead clearly presented as exactly that, L4S vs SCE.
>
>
>> This means declaring failure and say "no" to both SCE and L4S until
>> further notice.
> Given that SCE is not an adopted work this is more or less
> the defacto state for SCE, as it only gets WG time if the
> chairs so decide it is worthy of spending it.  On the other
> hand L4S has consumed a large amount of not only TSVWG time, but
> also of TCPM and ICCRG, and issues with it identified long ago
> have not progressed.  Is it time to put L4S to bed as a failure?
> Perhaps, it claims to have a large group of supporters, yet the work
> on it seems to be proceeding at glacial speed, and only by a few
> visible people.
>> I think both L4S and SCE has problems and I think proponents for both are
>> glossing over some of these problems.
> I do not believe that SCE has glossed over any problems,
> when we see one we have noted it and infact gone to great
> efforts to find problems, characterize them and seek solutions.
> It has been often that during that process we have identified
> problems with L4S, characterized them and provided feedback
> to the WG on what we have found.
>
> We are well aware of the defects in SCE and continue to work
> on resolving them.  We fully admit that SCE, as it is documented
> today, is not ready for deployment at scale, BUT we argue at least
> it is SAFE to deploy at scale.
>
> I would ask that you please identify a problem that SCE
> has glossed over, so that we might make certain it is on
> our known issues list and that we might address it.
>
>> What happened to the L4S issue # 16 discussion? It seems to have died in
>> June with no conclusion?
> Someone else in the thread has commented on this.
>
>> -- 
>> Mikael Abrahamsson    email: [e-mail]
Gorry