RE: voting membership rules
Mark,
My comments are inserted below.
-----Original Message-----
From: Roger B. Marks [mailto:marks@boulder.nist.gov]
Subject: voting membership rules
Gentlemen :
and Lady
I am a rookie Working Group Chair in need of some advice.
In constructing rules for 802.16, my greatest challenge is voting rights. I
simply can't decipher the 802 rules on this. Several of us spent over an
hour with Jim Carlo in Montreal without resolution. I've been worried that I
am overcomplicating the situation, but I have concluded that the situation
in inherently complicated.
I have looked at some other WG rules and have not found a clearer
explanation. I'd like to know more about how you interpret the rules in your
group.
Here are the key 802 statements:
"... Thereafter, voting membership in a Working Group is established by
participating in the meetings of the Working Group at two out of the last
four Plenary sessions... Membership starts at the third Plenary session
attended by the participant. One duly constituted interim Working Group or
task group meeting may be substituted for the Working Group meetings at one
of the two Plenary sessions."
"Membership is retained by participating in at least two of the last four
Plenary session meetings. One duly constituted interim Working Group meeting
may be substituted for one of the two Plenary meetings."
PAT: I agree with the material I deleted.
One scenario that also follows from the rules is:
Meeting: 1 2 3
Attendance: x x - x=attendance
Status: v v=becomes voter; n=becomes nonvoter
PAT: Note that the text says explicitly "Membership starts at the third
Plenary session _attended_ by the participant." Therefore, this person
would not become a voter at meeting 3 whether or not you had the
pettitioning requirement. We had a long discussion on this because the
phrase "attended by the participant" was not in there making the text
ambiguous. After discussing the pros and cons, we amended the text to add
that phrase to clarify. In chairing a working group, I would prepare a
slide before each plenary with the names of those who had participated in 2
meetings and were eligible to attain voting status at that plenary. Usually
less than 25% of the people on that list would attend a third meeting.
Geoff does the same in 802.3 and that applies. For a working group on a
"hot" industry topic, many people attend a couple of meetings to see what is
going on or because of interest in a transient issue. If one makes them
voters, one may end up with a lot people on the voting list who do not have
a continuing interest. (Definition of "a lot": at some meetings the list of
potential voters for 802.3 has filled the better part of a page with 3
columns of names.) There are letter ballots that you need a 50% return to
complete, so you don't want your voter list to get cluttered with
disinterested people. Three meetings seems a good filter.
Voting membership confers responsibilities as well as rights. In
priniciple, it is not right to obligate a person to these responsibilities
without the person indicating a willingness to assume them. That is why I
have always required that a person indicate a willingness to assume those
responsibilities by asking to become a voter before making them a voter.
There have also been a small number of people over the years who worked for
employers (usually government) that did not allow them to vote. If you make
them a voter, then they abstain or don't vote so it is better to not force
them into voting status.
In other words, you become a voting member at the third plenary even if you
don't attend it. I think the requirement that people petition for
membership at the meeting is in conflict with this rule, so I don't plan to
implement this petitioning requirement.
(2) Now we introduce the interim meetings, and things get trickier. The
problem is that the rules don't specify WHICH interim meetings are eligible.
For example, if someone comes to an interim in 1981 and then turns up this
November, does he become a voter the next time he shows up at a Plenary? A
more typical example is this: a guy comes to a March Plenary and a May
interim. Is he a voter in July? Does this violate the clause that
"Membership starts at the third Plenary"? It seems to; you could establish
voting membership from scratch in 4 months. How do you guys handle this?
One of the rules I'm considering is allowing an interim to substitute ONLY
for the preceding Plenary. This would require a minimum of 6 months to gain
voting rights. If I don't do this, I'll probably let the interim credit be
applied to either the preceding or following Plenary but not to any other.
PAT: My rule was always that the interim had to occur during the window
between four plenaries ago and the current plenary. If they use an interim
and a plenary, then I would require the plenary to be one of the last 3.
Otherwise, the person would fail the test of having attended "2 of the last
4 plenaries" at the end of the plenary at which voting rights were granted
and immediately loose their voting right which would not make sense.
The point of voting requirements is primarily:
ensure that voters are knowledgeble both about the work before the group
and its procedures
demonstrate the potential voter is likely to actively participate and
fulfill their obligations
A secondary point is to increase the difficulty/cost of "packing the room"
for a critical vote. The chair has other mechanisms to deal with block
voting and generally the 4 months it takes to get voting rights is long
enough to discourage it. I believe that someone who has an interest should
be able to become a voter in the minimum time to achieve the primary goal.
In my experience, someone who attends a plenary, the following interim and
the next plenary is generally an interested and active participant who is
able to vote knowledgeably and that 4 months in our fast moving industry is
not too short a time.
(3) This is a comment, not a question: I think that the rules should be
revised to take into account the existence of and importance of interim
meetings. Like many other groups, we are planning three interims a year.
People can maintain membership by attending two out of four plenaries, which
is three meetings every two years. Three out of twelve, in my opinion, is
insufficient to justify continued voting rights .
PAT: Since some groups in the past have held more than 3 interims a year,
requiring a level of interim attendence is a strain. Also, as a group
becomes more mature, there will be voters who care about maintaing the
integrity of the base standard who don't feel a need to be extremely active
on the supplement that is currently under development. Therefore, I believe
it is better to only require attendence at plenaries for voting rights.
There are proven means for making progress at interims.
The voting rights rule reduces the incentive for people to attend interims.
At our interim last week, we ended up with less than a quorum. It didn't
hurt us much, but it could in the future. For instance, my project plan has
us making our key decisions at a May 2000 interim. If we don't have a
quorum, we could have real problems.
I'm getting off the topic, but I'd appreciate any advice on how I can keep
from being completely hosed if I don't have a quorum. Right now, I have two
ideas:
-Make decisions by letter ballot.
-Get the inactive voting members off the rolls by:
-deleting members who fail to vote in letter ballots.
-offering inactive members the option to resign.
-ensuring that the rules are interpreted to delete inactive members. See
(4) below:
PAT: Your second measure is good. I would add:
Increase the likelyhood that someone obtaining voting rights intends to be
active which means require that they petition for voting status rather than
granting it automatically based on attendance.
The most powerful tool for making progress at interims is for the working
group to delegate the creation of drafts to task forces. 802.1, .3 and .12
have successfully used this for many years. So, at a plenary, you have a
vote to form one or more task forces, for instance a Physical Layer task
force and a MAC task force. Best practice is to also have an agreed on list
of objectives for the standard so that they have a yardstick to judge
proposals against and are most likely to produce something acceptable to the
plenary. 802.1, 802.12 and most 802.3 task forces don't have set membership
and have allowed voting by any attendee who feels sufficiently
knowledgeable. Therefore, there is no quorum for the task force. (There
was one task force in 802.3 on a very critical project that required
extremely detailed knowledge that did require attendance to establish voting
rights.) As things progess, there can be votes to authorize finishing a
draft to prepare it for ballot after the interim; review and respond to
ballot comments updating the draft based on the results, etc. You can look
at 802.3 plenary minutes to see how this works (the motions are usually on
Thursday).
Since 1985, 802.3 has only once had an interim 802.3 meeting (with quorum).
All other meetings were task force meetings and 802.3 has made rapid
progress.
(4) For a new WG, 802 doesn't include any specific rules except that:
"All persons participating in the initial meeting of the Working Group
become voting members of the Working Group."
Strictly interpreted, the rules says that my voting members (who became so
by attending last July) will lose their voting rights at the end of the
November plenary if they don't attend; they will not have attended two of
the last four. Of course, there have only been two, but the rules don't
provide any kind of allowance for that. One might say that one of two is
enough, given that there have only _been_ two. However, I prefer the
stricter interpretation and plan to use it. Note that people who lose voting
rights after November can regain it fairly quickly:
PAT: Sometimes a lot of people attend the first plenary of a new group
because they are curious or may be interested. Therefore, I think it is
desireable to expect the continued interest shown by attending the next
plenary to retain voting rights. The downside is it is possible a genuinely
interested and active voter could have an emergency that prevents
attendence. Fortunately, as chair, you have discretion to continue that
person's voting right. As chair for many years, I once got a request from a
person to extend his voting right because a medical condition made him
unable to attend which I granted.
Meeting: 1 2 3 4
Attendance: x - x - x=attendance
Status: v n v v=becomes voter; n=becomes nonvoter
I know you have higher-priority issues on your table, but I'd appreciate any
thoughts you can offer.
Roger
Dr. Roger B. Marks <mailto:marks@nist.gov>
Chair, IEEE 802.16 Working Group on Broadband Wireless Access
National Wireless Electronic Systems Testbed (N-WEST)
<http://nwest.nist.gov>
National Institute of Standards and Technology/Boulder, CO
phone: 1-303-497-3037 fax: 1-303-497-7828