ICANN/DNSO
DNSO Mailling lists archives

[wg-review]


<<< Chronological Index >>>    <<< Thread Index >>>

[wg-review] My comments


It's late on the 15th, so I'm sending comments. I'll frame them in terms of the
original post giving terms of reference.

First off, Karl Aurebach made comments early in the discussion that I agree
with fairly strongly:

(Karl's post is indented here, my additions not.)

       > 1. Objectives of the DNSO Review Working Group
       >
       > The DNSO Review Working Group's objective is to evaluate
       > the performance of ICANN's DNSO and to propose structural
       > and procedural changes that will help ICANN's Domain Name
       > Supporting Organization fulfill its mission of becoming a bottom-up
       > policy coordination body.
                ^^^^^^^^^^^^
       I have a bit of trouble with the limited powers implied by the word
       "coordination".

       The DNSO as a body is responsible not merely for "coordination" but also
       for the *origination* of policy pertaining to DNS.  The initiative for
       such policy might come from within the DNSO itself, by unsolicited input
       from the net community, or by reference from the Board of Directors or a
       question from another SO.

This is an important point.

       > The DNSO Review Working Group's objective is to evaluate
       > the responses of DNSO stakeholders' and to vindicate that DNSO
       > would be a structure that will include all of those who will be affected
       > by the DNS of the future as well as the current Netizens.

ICANN is widely perceived as oligarchic, elitist, dominated by the registries and
big business, unresponsive to small players. If fact, anything but the description
above, including "all of those who will be affected by the DNS of the future ...".

I think the problem is genuine, though it is often perceived as being far worse
than it actually is.

As I see it, the central problem for any review is how to open the organisation
up.

       We ought to dispense the concept of "stakeholders"

and its derivative, the constituency structure.

This was discussed extensively on the list. I do not think a consensus was achieved,
but there was certainly widespread feeling that the current constituency structure
was part of the problem.

Suggested solutions varied widely. Several people, including me, argued that
the whole constituency structure should be scrapped. On this topic, I quoted
one of my own posts to the GA list:

|> > The constituency structure is a failure and should be abandoned.
|> 
|> I agree, however could you suggest a better model, or how we could get
|> the constituency structure replaced and what it should be replaced by?
|
|The only model I know of that demonstrably works for the Internet is
|the IETF model. All working groups have mailing lists. Both those lists
|and the meetings are open to all comers.
|
|Moreover, everyone is expected to "leave their affiliations at the door"
|for working group meetings. The goal is to solve problems for the net,
|based on technical merit, NOT to represent your employer or interest
|group. Of course, it isn't actually that simple but at least the notion
|of leaving affiliations out of it is considered.
|
|This is almost the exact opposite of the constituency approach.

Others were more concerned with adding constituencies -- for individuals,
for domain name holders, for consumer advocacy groups, ... -- to remedy
perceived imbalances in the current structure.

As I see it, we should scrap the constituency structure. However, if we're
stuck with such a structure, then we should add constituencies to improve
the balance. At a minimum, constituencies for:

	end users, who after all are the final customer
	domain holders, if for no other reason than that they pay the bills
	consumer groups, to counterbalance the heavy business interests 

Continuing Karl's message:

       - particularly as some
       consider that concept to be one of the reasons why the DNSO is stumbling.

Yes.

       The DNS impacts everyone on the Internet.  That first sentence should be
       reworded to begin "The DNSO Review Working Group's objective is to
       evaluate the responses of interested persons" ...

       [SNIP]

       > * The DNSO constituency Structure : Examine the structure and
       >    propose amendments that will ensure balanced representation
       >    of all stakeholder interests in an open, and transparent process.
                   ^^^^^^^^^^^
       ...
       > In the long term, DNSO Review Committee will be responsible for
       > enhancing more trustworthy working environment in the DNSO
       > and for ensuring all the stakeholders' voices should be HEARD.
                                  ^^^^^^^^^^^^

       Again, that loaded word "stakeholders" - we ought not to pre-judge who has
       a "stake" but rather let people decide for themselves whether they feel
       that they have an interest they want to protect.  Rather than forcing
       people into pre-conceived, and arbitrary "constituencies" we ought to
       allow people to aggregate (and de-aggregate) into fluid coalitions.

Yes, indeed.

So those are my comments on Karl's comments. Then I have some of my own,
which I'll intersperse into (a heavily snipped version of) the terms of
reference post from the list archive.
  
       Name of WG : DNSO Review Working Group
       ================================
       Terms of Reference. version 0.3

[SNIP]

       These are issue list Review WG aims at making recommendations
       after its debates and discussion which will ameliorate DNSO/ICANN.

       * The role of working groups in the bottoms-up consensus process

That sentence has the best typo I've seen this year.

       * The Names Council's functions and responsibilities
       * The General Assembly's function and responsibilities
       * Re-examine the relationship between NC and General Assembly
       * Relationship between NC and ICANN staff : Better identify 
           which issues should begin with in the DNSO and which should 
           be handled by the ICANN staff. i.e.

If these issues were discussed, I missed it in all the noise.

The GA did come up, but mainly as a subject for argument over whether it
made an individual constituency unnecssary. Methinks it clearly doesn't.

       * Define a better procedure for forming working groups and for 
           making working groups productive.

Certainly working groups should (as in the IETF) be open to all comers
and should do most of their work on public mailing lists.

There was a long complex discussion of the term "consensus" and a widespread
feeling that ICANN misuses the term. Among the positions advocated were:

   "consensus" is just a buzz-word used to disguise rule by the chair.
     By declaring a bogus consensus, the ICANN cabal avoids issues and
     manipulates decisions.
   (Methinks there's a grain of truth in that one. Personally, I'm not
     certain there's any more than that, but a number of people on the
     list seemed to feel this was a major problem.)  

   "consensus" has a fairly precise meaning, involving either unanimous
     consent or "Quaker consensus" (everyone either consents or abstains,
     no-one feels strongly enough to block it). If that is not what we
     mean, we should drop the term. In particular, calling a 2/3 majority
     "consensus" (which some of the ICANN docs do) is both misleading and
     counter-productive. Also, it appears deliberately dishonest, so it
     is very much counter to the goal of openness.
   (I certainly agree with this position, as far as getting the terminology
     right. ICANN clearly does not run by consensus and should stop misuing
     the term. I suspect what we actually need is not a pure consensus model,
     but some midified form.)

   "consensus" has quite a different meaning in the IETF "rough consensus
     and running code" context. ICANN procedures are based on that; the
     ideas from other consensus processes are irrelevant.
   (I consider this notion ludicrous. ICANN makes policy, not code. The
     problem is quite different and it is not at all clear that ICANN
     processes, as currently implemented, resemble those of the IETF.)

   consensus is entirely innappropriate, the majority (after we get our
     new constituencies :-) must rule
   (I consider this one ludicrous too. It is far from clear what group
     you would need a majority of. It doesn't matter what you decide if
     there's no technically feasible way to implement it, so implementers
     must have some form of veto somewhere in the process. ...)

There seemed to be widespread agreement that the consensus process could
(and therefore should) be improved. Training the WG chairs in consensus
processes, providing guidelines for participants, ... I certainly agree
with that. 

At the same time, there were quite few people arguing that the process
needs to be more formalised. I agree in principle, but the proposals
I saw for doing that seemed to consist largely of trying to define
"consensus" as 90% agreement or some such. This strikes me as nonsense. 

       * The DNSO constituency Structure : Examine the structure and 
          propose amendments that will ensure balanced representation 
          of all stakeholder interests in an open, and transparent process.  
          Amendments and changes to the existing structures and processes 
          will be developed and posted for comment and discussion. 

As Karl points out, the term "stakeholder" is dubious. It is far from
clear who it might or should describe. Moreover, the notion of "balance"
here is almost equally dubious.  

       Relations between NC Review TF and Review Working Group

       In the short term, DNSO Review Committee is expected to review 
       the DNSO's responsibilities and its works and to approve(or modify) 
       the charter of Working Group F: DNSO Review and get the WG F
       formed and induce a wider debate.

Is there a draft of that chater somewhere? I haven't found it.
--
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



<<< Chronological Index >>>    <<< Thread Index >>>