Feedback-ID: i1136489e:Fastmail
Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3863.100.1\))
From: Edward Lemon <mellon@fugue.com>
In-Reply-To: <713a1f34-29ed-4fae-8113-62831a93b3dd@lear.ch>
Date: Thu, 31 Jul 2025 10:20:19 +0200
Content-Transfer-Encoding: quoted-printable
Message-Id: <D5F812F5-873B-46DF-B9D7-79FDD79D5574@fugue.com>
References: 
 <175377355078.585.16765936994715144258@dt-datatracker-5bd446d5fd-c47nq>
 <9ee7b233-f4b8-493f-b3e0-1b61b1743dd9@lear.ch>
 <F9E2CB7F-3232-4977-B45A-75C5FEBD21C7@fugue.com>
 <5003507a-f13b-4bba-914f-798a89ac3c8b@lear.ch>
 <D142D756-FAD5-4813-9554-83BAB822053E@fugue.com>
 <1a42c866-aa6d-413f-9a0f-49cb0ad7a515@lear.ch>
 <6A552735-A42D-4AD1-A4FE-EB19D83B7FA7@fugue.com>
 <D9605B34-D320-4E7D-8D91-D41CA488DD8D@episteme.net>
 <1b4d3a8a-5ef2-4d20-b3ac-7eaa0bdfc020@lear.ch>
 <8BE1758A-2C37-4E24-A57F-FAE08747D465@fugue.com>
 <0aa2b75c-21fd-4625-a279-2dd5af1e3fc1@lear.ch>
 <EF497690-98BB-46F3-B1EE-6E8ABE2BE7E7@fugue.com>
 <713a1f34-29ed-4fae-8113-62831a93b3dd@lear.ch>
To: Eliot Lear <lear@lear.ch>
Message-ID-Hash: PZXJB2D5W6CSPLAYXVGQJSLL4CTIG6EF
CC: Pete Resnick <resnick@episteme.net>, mod-discuss@ietf.org
Precedence: list
Subject: =?utf-8?q?=5BMod-discuss=5D_Re=3A_I-D_Action=3A_draft-ietf-modpod-group-proc?=
 =?utf-8?q?esses-09=2Etxt?=
Archived-At: 
 <https://mailarchive.ietf.org/arch/msg/mod-discuss/HdEwArkbn_NAGdMQnfERT1Z5QQI>

On Jul 30, 2025, at 22:26, Eliot Lear <lear@lear.ch> wrote:
>=20
>=20
> Even that example presents some challenges, like could a bad actor try =
to misuse the moderation process to slow an emerging consensus?  Keep in =
mind that moderators are not expected to monitor most fora, so they =
won't have the context.  With this in mind, do you want to take another =
shot at the text?


I think this concern about bad actors slowing the consensus process is =
overblown. I've never actually seen anyone do that, but I can think of =
several cases where feedback loops caused so much noise that evaluating =
consensus became impossible and the chairs gave up, with results that =
I'm quite sure didn't reflect the consensus. I maybe wrong that they =
didn't=E2=80=94that's the point=E2=80=94but this is clearly a way that =
process is harmed at the IETF for which we already have running code =
that works well. I haven't seen the other attack in the wild, so it's =
hard to take it very seriously.

We are also putting a lot of faith in the process of the moderators =
coming up with policies and procedures and best practices, and I think =
this falls into that.

Anyway, maybe something like:

- In particular, moderators should give WG chairs every opportunity to =
manage what may be=20
- difficult and contentious debates within their groups.  Within the =
bounds of this principle,=20
- it is left to moderators' judgement to determine when they must act.

+ Moderators are expected to defer to WG chairs knowledge of working =
group discussions and therefore=20
+ prioritize intervention by chairs where possible. Within the bounds of =
this principle, it=20
+ is left to moderators' judgement to determine when they must act.

I realize that this probably still isn't quite right, but it's the =
direction I'd suggest going.