Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802SEC] +++EC Motion+++ Rules Change Ballot on Roll Call Votes


With regard to the rule change ballot let me summarize
your position as I understand it:

Your experience / belief is that Roll Call votes are
only done when there is a question of the previous vote.
Furthermore you believe that Roll Call votes could be
used as a method of delaying progress and you specifically
mentioned 20% as too low a threshold given the technical
bar requires >= 75% concensus and repeated use of the
roll call vote to delay progress.

The Rules Change as currently drafted does have provisions
for the chair to limit the number of roll calls so your
comment about multiple roll calls is already addressed.

I can certainly solve the 20% issue, and I received a
comment from a non SEC member suggesting using the same
 >25% rule. In other words, if technical concensus cannot
be achieved, then delay will occur anyway and raising
the Roll Call requirement to 25% will not create the
ability for a minority that could other not delay
technical concensus from causing delay.

Will you change your vote with that modification? It
appears to be the only comment you gave and I am willing
to resolve it and I suspect that others would be fine
with that change.

Secret ballots in my opinion are not in keeping with the
open nature of 802 standards development and I have
yet to find a situation where one was needed.

You have avoided discussing the contentious issue and
since this is not a secret ballot let us focus our
attention there. In my experience, Roll Call votes
are only ever done to force people to go "on the record"
and not as a general delaying tactic. I could go on about
some of the really excellent turtles I have seen at
other standards bodies that use Roberts Rules to the
fullest extent possible and causing incredible delay
without a single roll call vote. Stuart Kerry has
pointed out that dot11 *does* use Roll Call votes regularly
on vital issues. Clearly roll call votes are not an
impediment in a group with a membership of over 500 and
several very successful standards.

The chair of a working group has a responsibility for
progressing the work of the group and for insuring the
openness and fairness of the process. There is a specific
rule in our P&P concerning insuring the process is not
being dominated by an organization.

Robert Love told me that he explained to you and the
WG the reasons why the chair should allow Roll Call
votes as a protection of the process and of the impartiality
of the chair.

I faced situations in dot17 where there was a belief that
progress was being blocked by an organization (including
my own). Roll Call votes were invaluable for me as a chair
to prove or disprove the existance of a block vote. The
roll call vote cuts both ways. If there is not block vote
occuring, then it closes the issue. In the one case where
the block existed, it allowed me to discuss the issue with
the blocking organization and by the next meeting, everything
was resolved. Role call votes occured rarely, and did not delay
any meetings even though we had 150+ voters. In fact with a
little preplanning, I had roll call votes runs extremely quickly.

There have been serious allegations made about membership & voting
behavior in dot20. I for one am interested in your action
plan to resolve the various allegations that have been
made publically and I would be happy to give you the benefit
of my experience in dot17 to help you plan a strategy to
insure that the process is not only fair, but seen to be fair.


mike wrote:
> Mike,
> Regarding 802.20, voting is done with colored sticky-labeled tokens that are changed for each session. Voters raise the tokens and counters announce the results as the parts of the room are counted. If the count looks confused or very close, then people are asked to raise their tokens again for another count.
> Since the formation of the working group, I cannot remember any roll call vote request made to either the two co-vice chairs or the chair.
> There was a motion made to have a roll call vote in November. This written motion was made immediately following the vote on creating a roll call vote operating rule in 802.20. The vote on this motion occurred at 4:49PM in the Thursday meeting with a scheduled adjournment at 5:00PM. The group still had a next meeting planning item to cover including a requested short presentation by Bob Love. Since the motion was immediately made and time was very short, I let the group decide by token voting whether to go through a roll call. The group decided 44 to 25 not to have a roll call. For any further details, I refer people to the draft minutes of the November session which are posted on the 802.20 site.
> Regarding the Roll Call ballot, my past experience is roll call votes are done when there is a question of the previous voting. If a member of the body questions the count of a voice vote, or a quick hand raise vote or questions the voting rights of people with or without tokens. Using token voting and having people keep their hands raised during an methodical and accurate count usually takes care of any issues.
> If the vote will be particularly sensitive or potentially contentious, then I always prefer a written ballot. A written ballot allows people to more easily vote the individual views without peer or organizational pressure.
> My other experience with roll call votes is they can be used to run out the time on an agenda item. For example if a group of people do not want a particular agenda item to come to the floor, then roll call votes in large bodies are used to stall until the the time runs out much like debates that are not managed by the chair. Clearly allowing a small minority of people(20%) this opportunity is not in the spirit of consensus. Given the real objective is a standard approved by consensus - - at least 75%, doing multiple roll call votes will not add value to the process.
> Additionally, I apologize in advance for not checking with Stuart Kerry on his current rules. However, I remember seeing in the minutes of the March 2002 802.11 Plenary a statement that roll call voting was impractical and discontinued given a current voting membership of 291. Again 802.11 uses voting tokens.
> Therefore, I currently vote disapprove on the proposed rules change.
> Regards,
> Jerry Upton
> Chair, IEEE 802.20
> In a message dated 11/25/2003 5:21:17 PM Eastern Standard Time, writes:
>>thanks for the improvement, I am happy to add
>>the obvious language fix and will think how to
>>integrate the additional instructions on
>>Between your examples of 1) and 2), I find 2)
>>to be the more problematic issue and the one
>>requiring attention since I believe Roberts
>>provides the chair with the necessary tools
>>to fix 1) without an explicit rules change.
>>I will look it up so I can cite the relevant
>>In particular, I have forced votes to be held
>>again when I failed to recognize a voter and then
>>reiterated that only voters may vote and the count
>>was different.
>>So let us discuss 2).
>>For example, in dot17, roll call votes were
>>used on several occasions to see if LMSC
>>rule e) was being violated. In 1 such
>>case, it was clear that the rule was being
>>violated and this allowed me to sit the
>>the entities involved down and read them the riot
>>act. Problem solved by next session!
>>The total number of roll call votes in 17
>>was low and there was only one occasion
>>where I thought the request for a roll call vote
>>was spurious and left it to a group vote, otherwise
>>I granted the request. In that one case, my recollection
>>is that it would not have hit the 20% mark. I am more
>>than happy to have my hands tied and hold such a vote
>>in the future.
>>As to the need for this in other working groups.
>>I said in the EC meeting that I am acting on a concern
>>from my former vice chair (Mr. Love) about the use
>>(or lack thereof) of roll call votes in dot20. I leave
>>it to him and other members of dot20 to highlight
>>why they see the need for such a rule.
>>I see the main use of a roll call vote is to get people to go
>>on the record and not so much to delay the process. As
>>I am sure some people will point out, there are better
>>procedural methods for delay. However, since *some*
>>people were concerned with roll call votes being used
>>to stall, I put in a remedy for that situation.
>>Personally, I can't imagine why a chair would refuse
>>to allow a roll call vote except to keep things
>>moving along. I can imagine why a group of voters
>>might want to avoid a roll call vote. It is in the vital
>>interest of IEEE 802 (and IMO IEEE) to insure that
>>people do go on the record for issues of importance
>>to the WG. Furthermore, as my failed EC motion shows
>>I believe this change should be implemented prior to the
>>start of the March 04 meeting.
>>Geoff Thompson wrote:
>>>(DISAPPROVE at this point)
>>>I applaud your attempts to fix this. I am not sure that it is a 
>>>tractable problem for reasons that I will outline below.
>>>But first the trivial fix:
>>>Change from:
>>>A roll call vote can be held at the discretion of the chair.
>>>Change to:
>>>    A roll call vote may be held at the discretion of the chair.
>>>And you need to add that a motion and a second is needed to call for a 
>>>roll call
>>>(and, I would hope that the vote could not be by acclamation)
>>>But further than that, I'm not wild about your test criteria.
>>>I believe that the problem in the Exec is that a roll call represents 
>>>two different situations.
>>>Situation #1
>>>Occurs in fairly large groups without voting tokens.
>>>Occasional problems arise when non-voters participate in a vote.
>>>There are two remedies to this situation
>>>        a) A caution from the WG Chair and a re-vote.
>>>        b) A roll call vote
>>>Requests for a roll call vote from the floor have traditionally been 
>>>Requests for a roll call vote have been very infrequent.
>>>Situation #2
>>>(This is somewhat speculative since I am not in these groups. I welcome 
>>>supplemental or corrective information)
>>>Occurs in groups with voting tokens
>>>These groups have less of a problem with non-voters voting.
>>>There are situations where there are strategic voting issues with voting 
>>>in a secret ballot vs a "public" roll call vote.
>>>A request for a roll-call vote can be used in this situation to slow 
>>>things down or to force members to vote "on the record".
>>>My feeling is that we need to (1) recognize these differences and (2) 
>>>come up with a solution that harmonizes or clearly differentiates these 
>>>2 situations rather than trying to lump them into the same pot.
>>>At this point I don't have a particular idea on how to do that, but I 
>>>believe that is the place to start.
>>>At 11:33 PM 11/24/2003 -0500, Mike Takefman wrote:
>>>>Dear EC Members,
>>>>as per the motion at the November Plenary closing
>>>>EC meeting I am starting a (35 day)  ballot on
>>>>the proposed rule change. I am extending the ballot
>>>>to account for the upcoming US Thanksgiving holiday
>>>>(and yes Canada has such a holiday - its just a month
>>>>I will be running a face to face comment resolution session
>>>>during the January Interim Session to try to finalize
>>>>the language. I believe sunday night is the best time
>>>>to hold such a meeting, but I am open to other suggestions.
>>>>The language you will find enclosed is different (and
>>>>I believe improved) from what was shown at the EC meeting.
>>>>1) It attempts to provide better sentence structure
>>>>(less of a run-on sentence).
>>>>2) It addresses an issue brought up to me personally
>>>>by one of the 2 dissenting voters to the rules change
>>>>motion in terms of insuring that roll call votes cannot
>>>>be used as a delaying tactic.
>>>>Personally, I have only seen roll call votes used in dot17
>>>>sparingly and they have in fact helped me determine when a group
>>>>was attempting to block concensus / progress. As such, 
>>>>has never been an issue with their use as a delay tactic,
>>>>but I do have sympathy for such a concern.
>>>>Michael Takefman    
>>>>Distinguished Engineer,       Cisco Systems
>>>>Chair IEEE 802.17 Stds WG
>>>>3000 Innovation Dr, Ottawa, Canada, K2K 3E8
>>>>voice: 613-254-3399       cell:613-220-6991
>>Michael Takefman    
>>Distinguished Engineer,       Cisco Systems
>>Chair IEEE 802.17 Stds WG
>>3000 Innovation Dr, Ottawa, Canada, K2K 3E8
>>voice: 613-254-3399       cell:613-220-6991

Michael Takefman    
Distinguished Engineer,       Cisco Systems
Chair IEEE 802.17 Stds WG
3000 Innovation Dr, Ottawa, Canada, K2K 3E8
voice: 613-254-3399       cell:613-220-6991