[wg-review] Preliminary Report - Consensus

[Greg Burton <sidna@feedwriter.com>]

Preliminary Report on Consensus, WG-Review
This is a preliminary report, and has not been reviewed by the membership 
before submission. While the poll material is public and reflects as much 
objectivity as possible, the opinions and conclusions drawn are the views 
of the author, and cannot be called "the results of the group" until the 
group has discussed and ratified them and any changes that may be incorporated.

Material comes from group discussions and two polls - one run through 
pollcat.com, and one run in a voting booth  designed for more secure results.

Problem Scope:
Asked "is the DNSO in need of reform", 36 out of 37 respondents agreed. It 
seems clear that among those responding, the overwhelming sense is that 
serious work is needed on the entire DNSO structure. In wide-ranging 
discussions certain themes emerged, and the question of "consensus" - what 
it means, what it is, and how it works - was one of those themes.

Over the course of this working group, it has become clear to group members 
that the entire issue of consensus is not only ambiguous, but is 
emotionally charged as well. Through group discussion, it was observed that:

"The WG-Review has observed that reaching a consensus within the DNSO was 
basically hampered by the lack of
definition and therefore of common understanding of what a DNSO consensus 
is and how it is determined"

This statement has agreed to by 14 of 15 poll respondents. Consensus was 
explicitly blocked on this question. As discussions further developed, it 
became clear that the DNSO's process was not what is normally understood as 
a "consensus process", but based, rather, in a process of the IETF. When 
asked to agree or disagree with the statement "I understand what the IETF 
"rough-consensus" model is and believe it is an appropriate model for the 
DNSO" 12 out of 15 respondents disagreed.

It appears that one reason for use of the IETF model is a belief that true 
consensus is impossible. Responding to the statement "I do not believe that 
true consensus can be achieved within the DNSO process at this time" 13 of 
the 15  respondents agreed. Interestingly, neither of the two who disagreed 
wished to block consensus on the question. Accordingly, this could be 
labeled a consensus result under formal consensus procedures. Irony is left 
to the reader.

2/3 of respondents (10 of 15) believe that they have learned something 
about consensus process through this Working Group. By a margin of 9 to 7, 
they believe that "for most issues, a simple majority vote is sufficient" 
None of the six who disagreed were will willing to block consensus on this, 
though the rather slim margin would indicate that consensus has not been 
achieved on this issue.

Action Directives:
Happily, the group achieved true consensus on one statement - " Clearly 
defined and published methods for considering DNSO issues would improve the 
process."

By a margin of 14 to 2, respondents believe "The use of "consensus" in the 
bylaws is misleading. Either ICANN should develop a true consensus process 
or it should change the bylaws to reflect the actual process desired." This 
is a significant margin though it is not consensus. It strongly suggests 
that on at least this issue some of the changes required to improve DNSO 
functionality will require changes to the ICANN bylaws.

Seventy-five percent of respondents to a question about whether or not this 
working group should call it's results "consensus" favored calling a vote a 
vote. This supports the belief that most question can be handled through a 
voting procedure mentioned above.

9-5, a majority of respondents believe "Some form of task force be 
developed as a training ground in consensus for ICANN and the DNSO." By a 
slightly greater margin (10-4) they believe that " NC members and WG chairs 
should participate in consensus training before heading a consensus-process 
group". And by an 8-5 margin, they agree that "A professional 
consensus-development consultancy should be engaged to advise on the 
consensus process."

Conclusion:
It appears that if the DNSO is going to support a consensus process, 
several things will need to happen. The most pressing issue from a 
consensus-development perspective is the significant belief that consensus 
is not possible in the DNSO at this time. It is a truism of consensus 
development that consensus cannot be imposed, and this material suggests 
that for a consensus process to be effectively implemented, it is first 
necessary acknowledge the need for GA  and constituency member education, 
as well as the suggestions above for the NC.

While working towards consensus is desirable, the current state of the DNSO 
will require explicit votes on matters of substance until a far greater 
level of trust in the process exists. A start at building that trust could 
happen by adopting all of the above proposals.

Greg Burton

--
This message was passed to you via the wg-review@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe wg-review" in the body of the message).
Archives at http://www.dnso.org/archives.html