I think this one warrants a separate thread......
--On 13. januar 2005 10:06 -0500 John C Klensin <john-ietf(_at_)jck(_dot_)com>
I think this is acceptable given that we *also* have a recall
procedure. In other words, if the IAOC isn't responsive
to a clear message from a review that "you screwed up", then
we'd better make sure that a recall is initiated.
And the traditional IETF recall procedure is probably not right
for this case, perhaps for one of the reasons it has never been
used in the IETF. The recall procedure requires parsing the
advocates and opponents of a decision (or screw-up more
generally), figuring out who is specifically to blame, and then
trying to remove him or her in a context that also requires
balancing past and potential future good deeds and speculation
about whether a replacement would be better against the
continuation of the person involved. The parsing process, and
maybe the rest, may be nearly impossible in a body we have
directed to work by consensus and that doesn't record votes or
internal informal discussions.
The same parsing process, of course, applies to "tell it to the
Nomcom" only without the benefits of public discussion of the
source of the dissatisfaction.
So the comments I've made have deliberately been of the nature
of "replace them", i.e., "fire the IAOC" rather than "recall an
individual". If the IAOC is going to work as a body and make
consensus decisions, then they are collectively responsible for
And I think the document probably needs some sort of "fire the
IAOC" procedure that would take out everyone except the three
(?) ex-officio people.
The recall procedure of RFC 3777 are in fact rather short....
7. Member Recall
The following rules apply to the recall process. If necessary, a
paragraph discussing the interpretation of each rule is included.
1. At any time, at least 20 members of the IETF community, who are
qualified to be voting members of a nominating committee, may
request by signed petition (email is acceptable) to the Internet
Society President the recall of any sitting IAB or IESG member.
All individual and collective qualifications of nominating
committee eligibility are applicable, including that no more than
two signatories may have the same primary affiliation.
Each signature must include a full name, email address, and
primary company or organization affiliation.
The IETF Secretariat is responsible for confirming that each
signatory is qualified to be a voting member of a nominating
committee. A valid petition must be signed by at least 20
The petition must include a statement of justification for the
recall and all relevant and appropriate supporting documentation.
The petition and its signatories must be announced to the IETF
2. Internet Society President shall appoint a Recall Committee
The Internet Society President must not evaluate the recall
request. It is explicitly the responsibility of the IETF
community to evaluate the behavior of its leaders.
3. The recall committee is created according to the same rules as is
the nominating committee with the qualifications that both the
person being investigated and the parties requesting the recall
must not be a member of the recall committee in any capacity.
4. The recall committee operates according to the same rules as the
nominating committee with the qualification that there is no
5. The recall committee investigates the circumstances of the
justification for the recall and votes on its findings.
The investigation must include at least both an opportunity for
the member being recalled to present a written statement and
consultation with third parties.
6. A 3/4 majority of the members who vote on the question is
required for a recall.
7. If a sitting member is recalled the open position is to be filled
according to the mid-term vacancy rules.
There's actually not much there about the criteria for removing someone -
it seems that if 20 people can get their act together and demand a
resignation, and 10 randomly-picked-from-a-pool-of-volunteer people can get
together, listen to the person under recall and agree (with 3/4 majority)
that the IETF is better off with getting rid of that person, that's it.
The one thing that I agree sticks out is that the language of 3777 talks
about firing *one* person - in the case where the group is dysfunctional,
it may be better to take the group out, as you say.
(there are more issues, such as the ISOC president no longer being a
disinterested party .... but those are less major issues than whether we
open up the topic of firing procedures...)
Ietf mailing list