Re: v 1.2, IETF material

["Beast (Donald E. Eastlake, 3rd)" <[e-mail]>]

From:  Carl Malamud <[e-mail]>
>                 STRUCTURAL CHANGES IN THE IAB and IESG
>
>1. TWO-YEAR TERMS
>
>1.1 Members of the IAB and IESG will serve two-year terms.  The terms are
>staggered so approximately half of each body is selected each year.
>There are no term limits.

How come IAB terms are 2 instead of 3 years as I though was being
talked about?

>2. NOMINATING COMMITTEE
>
>2.1 Selection shall be by a nominating committee.  The nominating committee
>is formed for a particular round of selections and then disbands.  This
>is not a standing committee.
>
>2.2 The nominating committee is composed of seven members selected by
>lot from a list of volunteers who are willing to serve and have
>attended two or more IETFs.  [Details on this procedure to follow.]

Should require that the IETF attendance be "recent".  Would someone know
what is going on now and what is required if they had only attended IETF
meeting >5 years ago?  How about adding "within the past two years at the
time of selection"?

>2.3 The chair of the nominating committee shall be the Executive
>Director of the Internet Society.   The IAB and the IESG may each
>appoint one liaison.  The chair and the liaisons are
>non-voting positions.

I do think that members of the IAB and IESG should not be on the
nominating committee except as the liaisons.  While you are at it, you
might as well specify that the nominations committee can't nominate
any of its own members.  This means that the IAB and IESG reps would
have to not be at the end of their terms which I think it good.  It
seems kind of embarrassing to me if the nominations committee had to
nominate someone who sat in on all its sessions.

>2.4 The nominating committee is responsible for recruiting nominees who
>are willing and able to serve.
>
>2.5 The nominating committee may recommend a single name for a
>position, or, if unable to reach consensus on a single name, multiple
>names.

I don't think the above section is needed at all.  At most, something
saying "The nominating committee shall recommend at least one name for
each position." should do and probably could be combined into 2.4.
Seems like it should just be up to the nominating committee how many
names they come up with.

>2.6 Selection of names shall be for a specific slot (e.g., "IAB Member"
>or "the Network Management Area Director of the IESG").  For purposes
>of this process, the Chair of the IETF is considered to be a specific
>slot.  The Chair of the IAB, however, is not a specific slot and that
>position shall be selected by the IAB from among its members.
>
>2.7 Both the IAB and the IESG may periodically change their composition
>to reflect the technical needs of the community.  If a new slot is
>created or an existing slot becomes vacant on the IAB, the IAB may
>appoint an acting person to fill that slot but that slot shall then be 
>subject to selection at the next regularly scheduled selection period.
>Likewise, the IESG may temporarily fill slots and those slots shall be 
>subject to selection at the next regularly scheduled selection period.
>
>2.8 The nominating committee may draw upon resources and advice from
>the community, but the committee shall keep all internal deliberations
>and personal data confidential.  The committee members shall agree in
>writing that the confidentiality of their deliberations are preserved.


>3. ADVICE AND CONSENT
>
>3.1 The selections of the nominating committee are subject to advice and
>consent.

I don't see any use in section 3.1.  Why not just delete it as its covered
in 3.2 and 3.3?

>3.2 Positions on the IAB are subject to the advice and consent of the
>Internet Society trustees.  The trustees may select one of the names
>presented to it or may reject all names.

If all names can be rejected, shouldn't it be said what happens?  Is
there a new nominating committee or does the body with the vacancy
just fill it until the next regular selection term rolls around?

>3.3 Positions on the IESG or the Chair of the IETF are subject to the
>advice and consent of the IAB.  The IAB may select one of the names
>presented to it or may reject all names.

Same comment as 3.2.

>3.4 Selectees shall give a talk at an IETF plenary soon after their
>selection.

Just to make it clear what is going on, I would add "to introduce
themselves" after the word "talk" in 3.4.

>4. OMBUDSMAN
>
>4.1 The Internet Society shall appoint a standing ombudsman to resolve
>conflicts.

You better specify that this person not be a trustee or a member of
the IAB or IESG but probably should require them to have attended two
IETF meetings within the two years before their appointment.  Does the
ombudsman have a term of office?  I think they should for just the
same reasons that IAB members, etc., do.  Could be 3 or 4 years.

>4.2 If a member of the IETF feels that a member of the IESG, the Chair
>of the IETF, or a member of the IAB is not carrying out his or her
>duties, the member of the IETF may present the situation to the
>Ombudsman.
>
>4.3 The Ombudsman shall attempt to resolve the situation within a
>90-day period.  The Ombudsman may gather the complaints of several
>people together into a single case, but shall always make a good
>faith attempt to hear all complaints.
>
>4.4 If the Ombudsman is unable to reach a resolution of the situation,
>a recall committee shall be formed to consider the case.

I have a problem with a possibly vague collection of verbal complaints
rolling along and picking up additional stuff.  (Well, I suppose that
a lot of this will be done by e-mail, which is arguably writing, but
some unclear subset of hundred of messages is just as bad as a verbal
hand wave.)  It someone is actually going to be recalled, I think it
is important to have a clear written statement of what the problem is
thought to be.  How about

"If the Ombudsman is unable to reach a resolution of the situation in
the above time limits, then a recall committee shall be formed to 
consider the case if any person complaining reduces their complaint to
writing and submits it to the Ombudsman."

Actually, I think calling on an Ombudsman appointed by the trustees to
intervene should always require a cohesive statement in writing as to
what the problem is so perhaps a similar provision should be inserted
in 4.2 which would eliminate a need for it here.

>4.5 The recall committee is formed by lot from a list of volunteers
>who are willing to serve and have attended 2 or more IETFs.

Same comment as above that attendance should be "recent", i.e., within
the previous two years.  Although I general don't like "definitions"
sections, you could solve this globally by adding something saying

"In these rules, the words volunteer and volunteers are limited to
those who (1) have actually volunteered their services in writing, (2)
have attended two or more IETF meetings within the previous two years,
and (3) are not Internet Society trustees or members of the IESG or
IAB."

(trivial nit:  2 should always be either 2 or two. This is currently
inconsistent.)

>4.6 The chair of the committee shall be the Executive Director of the
>Internet Society.  The IAB and the IESG may each appoint a liaison.
>The chair and the liaisons are non-voting positions.

It must be prohibited to appoint as a liaison someone complained
against.  (Unless the complaint is against the entire IAB or IESG in
which case it does not matter.  You might think this is silly but you
are setting up procedures where one person can file a complaint
against the whole IAB, for example, refuse to be satisfied by whatever
the Ombusdan does, and force a recall committee to be formed.  If its
a dillatory matter, the recall committee can just meet once and
dismiss the complaint, but you should be ready for this to happen.  If
you are ready for it, it probably won't happen. If you're not ready
and such screwing around will tie you in knots, then some idiot is
likely to try it.)

>4.7 The committee shall investigate the situation in a thorough manner
>and examine all relevant facts.

Seems to me that a one sentence or so of stuff here about the rights
of the person being complained against would be reasonable.  How about:
"The person whose recall is being considered shall be informed of the
membership of the recall committee and given a reasonable opportunity
to present their case to the recall committee."

>4.8 If the committee decides by a vote of 2/3 or more, a recall has
>been decided upon.  The committee shall publish a written document
>explaining its decision.
>
>4.9 The committee shall disband when it is finished deciding the
>specific situation it was formed to examine.  (This is not a standing
>committee.)
>
>4.10 If a member of the IAB or the IESG survives a recall motion, the
>Ombudsman shall not begin another procedure against that member for
>at least six months after the recall vote.

This is much too inflexible.  It should never be possible for someone
to become totally immunized to action against them.  Why not just make
the beginning of a procedure against a member after they survive a
recall motion be at the discretion of the ombudsman for the next 6
months or a year?

>5. TRANSITION
>
>5.1 This procedure will begin January 1, 1993 with selections to be in 
>place by the 26th IETF.
>
>5.2 The call for volunteers and selection of the nominating committee
>shall be done before December 31, 1992.
>
>5.3 The IESG will designate which positions shall be available for
>selection.  These positions will include the Chair of the IETF.
>
>5.4 The IAB will designate which positions shall be available for
>selection.  These positions will include the slot currently filled by
>the Chair of the IAB.
>
>5.5 The nominating committee will finish its work so that selections
>can be announced on March 1, 1993.
>
>5.6 The new selectees will take office at the close of the 26th IETF.
>
>6. STRUCTURAL CHANGES
>
>6.1 In addition to the institution of the selection and recall 
>procedures detailed in sections 1-5, the IAB and IESG will carry out
>a variety of structural changes in their internal operations and in
>the relationship between the two bodies.
>
>6.2 The IESG will institute the concept of design teams and will
>specify how those design teams relate to working groups.
>
>6.3 The IESG will make area directorates a standard part of the
>area concept.
>
>6.4 The IESG will make decisions on standards progression.  It will
>institute procedures that will bring more quality control towards
>the working group level, and will specify procedures for the progression
>of working group documents through the area directorates and the IESG.
>
>6.5 The IAB will have the role of arbitrating disputes and acting
>as an appellate body.

This is awefully vague.  If that's the best that can be done with the
current consensus, I don't want to derail things.  But if the idea is
that the IAB is a place where violations of process can be appealed,
shouldn't this rule just say so?

>6.6 The IAB will provide oversight and advice on the Internet architecture.
>The IAB will review working group charters and provide advice on 
>architectural issues that may be raised by those working groups.
>
>6.7 The IAB will continue to participate in the IETF process and may
>participate in working groups and issue proposals and other papers
>that contribute to that process.
>
>6.8 The IAB and the IESG will periodically report back to the IETF
>on the status of these changes.

Despite the extent of my comments above, I think this is a excellent
piece of work.

Donald