Re: please take care in tinkering

"James R. (Chuck) Davin" <davin@bellcore.com> Mon, 30 November 1992 22:14 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa16748; 30 Nov 92 17:14 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa16737; 30 Nov 92 17:14 EST
Received: from ietf.cnri.reston.va.us by CNRI.Reston.VA.US id aa17144; 30 Nov 92 17:15 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa16718; 30 Nov 92 17:14 EST
Received: from thumper.bellcore.com by CNRI.Reston.VA.US id aa17104; 30 Nov 92 17:14 EST
Received: from phila.bellcore.com by thumper.bellcore.com (4.1/4.7) id <AA19921> for poised@CNRI.Reston.VA.US; Mon, 30 Nov 92 17:15:01 EST
Received: from localhost.bellcore.com by phila.bellcore.com (4.1/4.7) id <AA06758> for poised@CNRI.Reston.VA.US; Mon, 30 Nov 92 17:15:20 EST
Message-Id: <9211302215.AA06758@phila.bellcore.com>
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: "James R. (Chuck) Davin" <davin@bellcore.com>
To: poised@CNRI.Reston.VA.US
Subject: Re: please take care in tinkering
In-Reply-To: Your message of Mon, 30 Nov 92 08:15:17 -0800. <9211301615.AA18912@aland.bbn.com>
Date: Mon, 30 Nov 1992 17:15:17 -0500
X-Orig-Sender: davin@phila.bellcore.com

I believe very strongly that the recall procedure is a good idea.  If
there are ever times where a change of direction is required, it is
unlikely that the community will be able (even if it were willing) to
wait through a normal cycle to effect that change. Instead, we will
have a crisis, and we won't have any timely way of resolving it.

I further share some of Craig's concerns about tinkering because,
while some have speculated about how painful a recall procedure could
be, others have at the same time engaged in tinkering the original
proposal to make it ever more painful. Someone was concerned that a
recall could take on the horror of a public crucifixion; at the same
time someone else is proposing that a recall begin with some highly
public call for information ("indictment"?) or for a new pool of
volunteer jurors ("partisans"?).

To be credible and useful, a recall mechanism must be easy to invoke,
but should by no means allow for casual dismissal of leaders, and
should by no means be unduly painful or humiliating.

Such a procdure is not hard to specify, and what has already been
presented is (modulo a few nits) quite workable.

When a recall committee is organized, there should NOT be a public
call for volunteers at that time. Both nomination and recall
committees should be drawn from a standing pool of volunteers, who
shoulder this responsibility independently of the particular future
circumstances in which they may be called to serve.

A recall proceeding should be a discrete, quiet affair. It is the
responsibility of the Ombudsman to *quietly* resolve any complaints on
an informal basis, and, if needed, to *quietly* form the committee by
lot, to *quietly* convene the committee and bring it to a speedy
conclusion.  I agree with a previous writer who said that the only
public output of a nominating or recall commitee should be a list of
nominees or an annoucement of removal, respectively.

How could we shield our leaders from "nuisance" recall actions without
making the recall procedure so expensive to invoke that it cannot
credibly fulfill its function of allowing timely changes of direction?

"Nuisance" recall actions are really a question of how painful or
inconvenient it is to discharge a clearly insubstantial complaint.  If
the recall procedure is not tinkered into being a matter of public
theatre and gory spectacle, then that keeps the cost of the procedure
low in both "legitimate" *and* "nuisance" cases.

But we may want the cost to be even lower in "nuisance" cases. For
this, add a few easy conventions that do not compromise the integrity
of the process:

1. Batching of complaints: Posit that recall committees need not meet
outside of IETF meetings. If some individual or lunatic fringe group
brings multiple complaints against the same leader or group of leaders
in the period between two IETF meetings, then the Ombudsman can choose
to batch these complaints and resolve them in one recall proceeding.

2. Because invoking the recall procedure should not require attendance
at an IETF meeting, it should be possible to present one's case to a
recall committee via email. For fairness, both complaintants and
accused leaders should be able to present their cases via email if
they so choose.

These two conventions taken together mean that the cost of "nuisance"
invocations of the recall procedure are quite low: when a complaint is
truly insubstantial, an accused leader can discharge it by making his
case in an email message that the Ombudsman presents to the recall
commitee. And this need happen no more often than once every IETF
meeting cycle. Admitting cases to be made via email through the
Ombudsman both reduces the cost of nuisance attacks and also assures
continued participation in the IETF for those who can't always attend
the meetings.

2 cents,
Chuck


P.S.  Nits on Carls' text:

Add text about batching of complaints and making recall cases via
email as sketched above?

Could say that when the ombudsman drafts (by lot) volunteers for a
recall committee, they should not be told about the case or who is
involved until they have agreed to serve.

In general, could state that the Ombudsman is responsible for
conducting the business of her office and the business of nominating
and recall committees in a private and discrete manner.

Does not explicitly say how many folk are on a recall committee. One
assumes 7, but it's not clear.

Says recall committee is chaired by ISOC Exec Dir. Why not the
Ombudsman instead? Are they assumed to be the same?