Justin's proposed changes.

[Michael Dillon]

On Sun, 26 Jan 1997, Justin W. Newton wrote:

> 2) No BoT or AC member can hold a paid position inside the non-profit.
> (There is too much conflict of interest there).  If we /have/ to have a BoT
> or AC member that works for the non-profit it should definately be a
> non-voting position.

I'm inclined to agree with this. I presume you are saying that the
executive director should have a permanent non-voting seat on the BoT?
I wonder how this would work out if the BoT were planning to make its
decisions by consensus rather than by vote. Unless there is some codified
system for determining presence or absence of consensus it is hard to have
a non-voting person present at the discussions.

Does anyone have any information about organizations that HAVE worked with
some sort of codified consensus decision making system? I can think of
possible ways to do this but unless they are tested elsewhere I'm not sure
the ARIN membership would like to see such a thing. For instance, there
could be 3 states of a vote rather than the traditional two, for and
against.

           I am for the motion.
           I am opposed to the motion but willing to go along.
           I am against the motion.

If those three states were weighted 1, 0.5 and 0 then one could say that
consensus was reached when the sum of the voting states was greater than 
50% of the number of members. Thus, on a board with 6 members, if 1 member
is opposed to a motion and all others are willing to go along we have
a voting status equal to 5 * 0.5 + 1 * 0 for a total of 2.5 which is
less than the total of 3.0 required. However, if one of those people
changed to support the motion the voting state would be
1 * 1 + 4 * 0.5 + 1 * 0 for a total of 3.0 which passes the motion.

I would consider the former example to be a lack of consensus but the
latter example to indicate consensus. However, at the other end of the
spectrum, a vote of 3 in favor and 3 opposed would also total 3.0 and I
would consider this state to indicate no consensus at all. Therefore I
would modify the above algorithm to say that consensus exists under one of
the following 2 conditions which don't require a weighted algorithm at
all:

    1) no one is against the motion

    2) 1 person is against the motion but at least one person
       supports the motion and all others either support the motion
       or are willing to go along.

Once consensus is codified by some scheme such as this, the presence of
non-voting members in the discussion cannot influence the presence of
consensus. This also allows for a record of who voted how because the
three voting states are discrete as opposed to a loose consensus system
in which there is no clear indication of how an individual voted.

I know some people will point to the IETF and the IETF credo here in
support of the system of loose consensus. However the IETF has the
"running code" proviso which places some limits on the loose consensus.
In the case of ARIN, the BoT has to interface to normal "majority rules"
entities such as the membership who elect the Advisory Council and may
even elect the BoT. Therefore I believe that if consensus is to be used to
make decisions within the BoT, it must be a codified form of consensus.

Michael Dillon                   -               Internet & ISP Consulting
Memra Software Inc.              -                  Fax: +1-250-546-3049
http://www.memra.com             -               E-mail: michael at memra.com