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

RE: [802SEC] RE: Rules Language interpretation on Voting




Buzz,

I don't recall the 802.1 ballot, but that isn't surprising because I was not
monitoring 802.1 closely in the early days. As I mentioned, block voting
becomes a more significant concern as group size gets smaller. 802.1 has at
times been fairly small. At one point it was under 10 people at a plenary
meeting and a number of them were from the same company. This does not imply
anything negative about that company. Its attendees were and are strong
contributors and that it maintained support for the work of 802.1 during the
part of the 80s when the interest of others was on the wane is to its
credit. (I don't think it is the same as the company to which you refer.)
But one can see that with that small a group even 3 well-intentioned
attendees from the same company can create concerns.

Regards,
Pat

-----Original Message-----
From: Rigsbee, Everett O [mailto:everett.o.rigsbee@boeing.com]
Sent: Monday, March 25, 2002 8:05 PM
To: 'pat_thaler@agilent.com'; tak@cisco.com
Cc: stds-802-sec@ieee.org
Subject: RE: [802SEC] RE: Rules Language interpretation on Voting


Pat,  Actually, what you described is what I thought I said:  that you take
both because the per company vote is not IEEE legal, but it does provide
useful data for comparison.  

And I do remember at least one time when the rule was actually used against
a large blue company that was silly enough to submit Xeroxed copies of the
same set of ballot comments on the ballot responses of several different
employees for an 802.1 letter ballot.  Those ballots were all counted as a
single vote.  


Thanx,  Buzz
Dr. Everett O. (Buzz) Rigsbee
Boeing SSG
PO Box 3707, M/S: 7M-FM
Seattle, WA  98124-2207
Ph:  (425) 865-2443
Fx:  (425) 865-6721
Email:  everett.o.rigsbee@boeing.com

-----Original Message-----
From: pat_thaler@agilent.com [mailto:pat_thaler@agilent.com]
Sent: Monday, March 25, 2002 5:13 PM
To: Rigsbee, Everett O; tak@cisco.com
Cc: stds-802-sec@ieee.org
Subject: RE: [802SEC] RE: Rules Language interpretation on Voting

Buzz,

I wouldn't call taking a company vote common practice. At no time did we
base a binding vote on a company vote. On a small number of occasions, we
did a vote taking tallies on one vote per voting member and on one vote per
comany (the latter vote being for information). The first time I did it was
actually during a 10BASE-T task force meeting (perhaps in November 1987)
when there was an important vote on whether to extend the window for new
proposals. (For background, the common voting practice in 802.3 task forces
is to allow all participants to vote. Whatever the taskforce decides has to
be accepted by 802.3 so that is where we apply a formal voter list and in
most task forces we follow a more informal rule.) As time for the vote
approached, an increasing number of people from a certain company which
wanted to extend the window flowed into the room - people who normally
attended .1, .4, and .5 or other .3 task forces. In response to concerns
raised from the floor that the room was being packed, we took the vote both
ways - individual and per company. The percentage vote was identical for
both tallies.

On the small number of other occasions when we have tried this exercise, we
have seen the same result.

I attribute this to two things: it is hard to pack the room to change the
voting when the group is large. One company may have a noticably large
number of people voting on one side of an issue but the may be a lot of
companies with two to four people voting on the other side and the company
with the large block may have a number of other companies with single voters
so that the situation balances out.

Block voting and packing the room are primarily problems for smaller groups.
I don't think we have ever invoked the rule for limiting a company's voters
because of block voting. It is pretty hard to prove conclusively unless it
is extremely blatant. Usually we have prevailed with more gentle methods
(similar to what you describe).

Regards,
Pat Thaler



-----Original Message-----
From: Rigsbee, Everett O [mailto:everett.o.rigsbee@boeing.com]
Sent: Monday, March 25, 2002 4:01 PM
To: 'Mike Takefman'
Cc: 802 Exec (E-mail)
Subject: [802SEC] RE: Rules Language interpretation on Voting



Mike,  I believe that it was a common practice in Pat Thaler's 802.12 group.

They frequently took both individual and per-company votes on the same
motion to see if there was a dramatic difference. 
When dealing with potential Block-Voting abuses, having that kind of data to
refer to helps a lot if you do have to make a ruling.  Roll call votes also
provide precise data. 
When folks are thus made explicitly aware of the impending hammer, they tend
to modify their behavior to mitigate the problem. 
Per company voting is not recognized by IEEE as a valid mechanism, so you do
need to do both, but the data collected can make a convincing case for
reducing all the votes from a single company to count as just a single vote
per the rules.  In Pat's case, the mechanism worked very well and the
problem soon went away.  You should ask her for more details. 


Thanx,  Buzz
Dr. Everett O. (Buzz) Rigsbee
Boeing SSG
PO Box 3707, M/S: 7M-FM
Seattle, WA  98124-2207
Ph:  (425) 865-2443
Fx:  (425) 865-6721
Email:  everett.o.rigsbee@boeing.com

-----Original Message-----
From: Mike Takefman [mailto:tak@cisco.com]
Sent: Thursday, March 21, 2002 2:11 PM
To: Tony Jeffree
Cc: Paul Nikolich; p.nikolich@ieee.org; thompson@ieee.org;
bob.grow@intel.com; Rigsbee, Everett O; bob@bstormnetworks.com;
stuart.kerry@philips.com; bheile@ieee.org; v.hayes@ieee.org; wlq@cisco.com
Subject: Re: Rules Language interpretation on Voting

Gentlemen,

thanks for the responses so far, I agree that descending into a
rules arguement in the WG will cause us to enter a rather special
level of hell that I want to avoid at almost all costs. I say almost
because if at the end of the day, this startup refuses to act
reasonably then the I have to decide if we want to just wait them
out, which I fear would put the market acceptance of the standard
at risk.

I have informed the two leaders of the companies involved that
the behavior was unacceptable and that they owe the WG simulation
results and presentations if they continue to block at the next
meeting. And their already sullied reputations will be further
tarnished (hows that for switching metaphors).

I do not like to threaten that particular rule if I cannot go
through with it, as a bluff called leaves you with much less
power in the future. It has been mentioned to me by one old timer
that per company votes have been done in the past although
rarely. Does anyone have direct experience with it? I.E. can the
WG make a proceedural decision to hold a vote on a 1 Vote per
company basis?

cheers,

mike

--
Michael Takefman              tak@cisco.com
Manager of Engineering,       Cisco Systems
Chair IEEE 802.17 Stds WG
2000 Innovation Dr, Ottawa, Canada, K2K 3E8
voice: 613-254-3399       fax: 613-254-4867