[Internal-cg] Consensus building process

Joseph Alhadeff joseph.alhadeff at oracle.com
Mon Aug 18 10:01:18 UTC 2014


As we consider consensus, I think we all agree that while we consider all inputs, consensus is across our working group that has representation from all the major stakeholder communities. We may need to make that clearer to level set expectations from folks or other groups that submit comment.  This will further reinforce the need and benefit to channeling comments through community processes.

Joe

Sent from my iPad

> On Aug 15, 2014, at 2:06 PM, Patrik Fältström <paf at frobbit.se> wrote:
> 
> All,
> 
> I find some of this discussion refreshing, and important, because we will most certainly end up in these kind of corner case situations one day or another. That said, I do not think we can envision all different cases.
> 
> Further, I think we have to separate the question on whether people agree or disagree to "the main proposal" from agreeing or disagreeing to "the conclusion of the discussion" (which might include information about minority views etc).
> 
> Yes, we are to deliver _one_ proposal, but even one proposal might include various wordings [based on minority views] that round off the corners so that the actual product from ICG might be acceptable, while "the main path" might not be.
> 
> Because of this, I am like Martin nervous over voting, because it will most certainly for many reasons be hard to vote about some text. Instead, we will have to work with multiple versions until our chair do declare the text be as good as it can be [I am not to declare or repeat what the basis should be for that decision].
> 
> Yes, it will be hard, but I do feel we do understand all of us we in this group want to get some deliveries. And not deadlock.
> 
> So I am optimistic.
> 
>   Patrik
> 
>> On 15 aug 2014, at 19:10, Martin Boyle <Martin.Boyle at nominet.org.uk> wrote:
>> 
>> Hi Joe,
>> 
>> I am concerned by the return to the idea of supermajorities overruling significantly affected parties.  I do not like recourse to voting (for reasons that I have already explained), and certainly not when it can be used to overrule legitimate concerns.
>> 
>> First, I agree with your division between non-substantial (those not related to the final proposal, but focussed on our ways of working, charter, even the RFP) and the substantial (related to proposals).
>> 
>> For your hyper-majority proposal (which reminds me of EU formulae for Council decisions – and that’s not an accolade!), what is the percentage?  I gave some thresholds in my original mail on this issue.  What happens if one of the dissenting organisations were to be the gTLDs?  Or the ccTLDs?  (Interestingly, I don’t think such a mess would happen in the case for numbers or protocols.)
>> 
>> And being voted against won’t solve the problem because you won’t have addressed the specific concerns of the beaten party who would then lobby against the solution showing that we do not have a consensus proposal.
>> 
>> Essentially, I see a distinct risk of a formulaic approach leading to a risk of marginalising particular interests simply because they don’t fit the model others might prefer.
>> 
>> I would note that the role of the ICG is to represent the wider community.  However, we are not elected by them, but appointed by our communities and accountability to those communities.  The chair’s role is to try to pull together consensus, ensuring that serious concerns are properly considered and that the output is something that works for all and where there are appropriate safeguards.  I agree that the discussion can’t keep going around in loops (hence the need for us to justify our positions) and sometimes there will be a final dissenting voice who does need to be given a say in the final document.
>> 
>> Cheers
>> 
>> Martin
>> 
>> 
>> From: internal-cg-bounces at icann.org [mailto:internal-cg-bounces at icann.org] On Behalf Of joseph alhadeff
>> Sent: 14 August 2014 20:07
>> To: internal-cg at icann.org
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Colleagues:
>> 
>> Perhaps we can consider a different dividing line on consensus.  I would suggest that we have two main buckets, that which is related to a proposal and that which is related to our operations.  For operations 2/3 of those expressing an opinion is sufficient to arrive at decisions, but for issues related to proposals we need more than 2/3 (other percent?) of all members with no more than 2 participating organizations dissenting. The role of the chair would thus be constrained by rules regarding when consensus is achieved, but the chair would have the ability to manage discussion or call for a formal vote if they believed it would help move us towards consensus.
>> 
>> Slightly less complex and perhaps more action-oriented.
>> 
>> Joe
>> On 8/14/2014 2:26 PM, James M. Bladel wrote:
>> Agree.  “Consensus” is not equivalent to “Unanimity.” 
>> 
>> J.
>> 
>> 
>> From: <Drazek>, Keith Drazek <kdrazek at verisign.com>
>> Date: Thursday, August 14, 2014 at 13:10 
>> To: ICG List <internal-cg at icann.org>
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Just so I’m clear…
>> 
>> Looking ahead….if we end up with 29 ICG reps in favor of a final recommendation and one person who unreasonably refuses to compromise, will that be deemed “consensus” or “no consensus?”
>> 
>> Hypothetically speaking, if one holdout among us can obstruct a decision that has received support from all other members, and would prevent delivery of a recommendation….I find that very troubling.
>> 
>> Keith
>> 
>> 
>> 
>> From: internal-cg-bounces at icann.org [mailto:internal-cg-bounces at icann.org] On Behalf Of WUKnoben
>> Sent: Thursday, August 14, 2014 1:47 PM
>> To: Kavouss Arasteh
>> Cc: Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Dear Kavouss,
>> 
>> you make the same point I expressed by saying that “I’m still uncertain with “non-substantive” issues which level of substance may depend on different views”. I would welcome you providing other more useful criteria to decide in which rare cases a “poll” or “voting” could apply.
>> 
>> As you may have seen in my latest draft I removed the “adjectives” from consensus. So I would appreciate your written suggestion for an acceptable text that I could better understand your disagreement with the present proposal.
>> 
>> Best regards
>> 
>> Wolf-Ulrich
>> 
>> 
>> From:Kavouss Arasteh
>> Sent: Thursday, August 14, 2014 7:22 PM
>> To: WUKnoben
>> Cc:Milton L Mueller ; Martin Boyle ; Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Dear All,
>> I am not comfortable to any of these measures.
>> The more we discuss and analyze ,the more problem is created.
>> I strongly disagree to make any discrimination among the contstituent groups in ICG ,WHEN IT IS PROPOSED qUOTE
>> "For substantive issues, at least none of the “customer groups” (numbers, protocols, gTLDs or ccTLDs) of the IANA remains strongly opposed"
>> What is considered by someone " substantive" may be considered by others " non substantive,
>> NO ADJECTIVE FOR OPPOSITION .NO ADJECTIVE FOR SCONSENSUS.
>> If you want instead of making progress to draft another chatter or convention for decision making ,I disagree with that .
>> It ios incumbent to the chair and the two vice chairs to make utmost efforts to build consensus-
>> Pls end this discussion
>> Regards
>> Kavouss
>> 
>> 
>> 
>> 
>> 2014-08-14 18:32 GMT+02:00 WUKnoben <wolf-ulrich.knoben at t-online.de>:
>> Thanks all for your valuable input.
>> 
>> Milton is right calling for verbal clarity. But differentation is also needed and there are different approaches to achieve it. And as I said before the suggestion so far was based on GNSO habit.
>> 
>> I tried to accomodate the discussion and therefore suggest to differentiate between “recommendation by consensus” (highest level, 100%) and “recommendation” (all remaining discussion results leading to a recommendation).
>> 
>> I agree to all basic principles Martin came up with and incorporated them.
>> I’m still uncertain with “non-substantive” issues which level of substance may depend on different views.
>> 
>> I would appreciate further fruitful discussion on the list and we will hopefully see an end at our call next week.
>> 
>> See my edits attached.
>> 
>> Best regards
>> 
>> Wolf-Ulrich
>> 
>> 
>> From:Milton L Mueller
>> Sent: Tuesday, August 12, 2014 8:12 PM
>> To: 'Martin Boyle' ; Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> I think Martin makes very good points here.
>> I like his proposed principles, every one of them.
>> 
>> I must confess that I have been wincing at the way the word “consensus” is (ab)used routinely in these circles. Either it is truly consensus, and everyone either agrees or agrees not to object, or it is _something else_. Will we please stop trying to apply the term “consensus” to supermajority voting processes? My academic commitment to verbal clarity and directness is screaming at me that this is wrong.
>> 
>> The IETF concept of “rough” consensus is an informal mechanism that is suitable for a more homogeneous environment in which adherence to standards is voluntary anyway, but in an environment with binding outcomes and political factions, it can and, in the ICANN context, frequently HAS merely provided a rationalization for ignoring significant minority points of view.
>> --MM
>> 
>> From:internal-cg-bounces at icann.org [mailto:internal-cg-bounces at icann.org] On Behalf Of Martin Boyle
>> Sent: Tuesday, August 12, 2014 1:24 PM
>> To: Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Hi All,
>> 
>> First thanks to Wolf-Ulrich for his paper.  I greatly like the idea of standards of good behaviour and mutual respect – and I’m pleased to see that this is already very much the framework for the way that the ICG works.  I’d also note that the analysis of shades of grey in levels of support is interesting – was it Patrik who first noted the two extremes (non-substantial and substantial issues) and the level of consensus that might be needed?  I’m just not sure I know how to use them…
>> 
>> I’d firmly endorse the aim that “the ICG … reach at least Consensus on the Proposal for the IANA Stewardship Transition to be forwarded to the NTIA” subject to our continued effort to try to achieve full/unanimous consensus or (at least) to have addressed address points of concern.
>> 
>> However, I do not like processes that are supposed to be by consensus being resolved by voting (cf WCIT):  voting leaves winners and losers.  It also means that people get lazy and fail to look for compromise or common ground or ways to address “reasonable” concerns.  That aversion is not really addressed by supermajorities:  even at an 80% supermajority, all the domain name registries or all the government representatives or all GNSO members could be overruled.  At 85% all the ccTLD registries, at 90% all the gTLD registries could be ignored.
>> 
>> I do recognise the need for a mechanism that allows us to come to a final recommendation and I’m afraid that I do not see any magic wand.  But I would suggest a number of basic principles:
>> 
>> ·         The aim of the discussion should be to try to find a solution where *no member of the ICG still maintains serious opposition to the outcome.*  Reasons for objections should be given, allowing the ICG wherever possible to try to address those concerns.
>> 
>> ·         *Recourse to any form of voting should be the exception.*  Its use might be fine for non-substantive issues.  For substantive issues, at least none of the “customer groups” (numbers, protocols, gTLDs or ccTLDs) of the IANA remains strongly opposed.
>> 
>> ·         Group members who still have problems with the evaluation should be invited to *identify possible ways in which the proposal could be modified to make it acceptable to them.*
>> 
>> ·         Discussions should continue until *no “IANA customer” group is firmly opposed.* 
>> 
>> 
>> One final point:  I would be willing to allow anyone who feels that they have not been heard to put a minority view into the final report.  I’d rather that did not happen, but if the views are strong enough, it would be best to have then documented in the report than to be first aired in the discussion that follows the publication of our final report.
>> 
>> Cheers
>> 
>> Martin
>> 
>> 
>> 
>> From:internal-cg-bounces at icann.org [mailto:internal-cg-bounces at icann.org] On Behalf Of Kavouss Arasteh
>> Sent: 11 August 2014 20:48
>> To: Drazek, Keith
>> Cc: Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Dear All,
>> Undoubtedly, it would be super majority either 2/3 or 4/5 .
>> Kavouss 
>> 
>> 
>> 2014-08-11 18:18 GMT+02:00 Drazek, Keith <kdrazek at verisign.com>:
>> I agree that we will need a clear process for determining consensus that falls somewhere on the spectrum between humming and requiring a unanimous vote.
>> 
>> If we get in to discussions of voting, we'll also need to address the thresholds required to establish consensus. Is it a simple majority? Super-majority?  Unanimous voting is an unhelpful requirement that would likely obstruct our work and our ability to deliver, so I believe that should be a non-starter for the ICG. We need to avoid the possibility of one dissenting vote undermining an otherwise strongly supported recommendation that represents broad community consensus.
>> 
>> However, if/when there is not full consensus, it will be important that we have a mechanism for expressing dissenting opinions. The GNSO Registries Stakeholder Group employs a "minority statement" mechanism to allow for all views to be expressed when there is consensus but not unanimity on a particular topic. Perhaps we should consider a similar mechanism for the ICG.
>> 
>> Keith
>> 
>> -----Original Message-----
>> From: internal-cg-bounces at icann.org [mailto:internal-cg-bounces at icann.org] On Behalf Of Subrenat, Jean-Jacques
>> Sent: Monday, August 11, 2014 6:09 AM
>> To: Kavouss Arasteh
>> Cc: Coordination Group
>> Subject: Re: [Internal-cg] Consensus building process
>> 
>> Hello Colleagues,
>> 
>> From the experience of the past few weeks, unfortunately we can conclude that the current process is not successful. Rather than meting out blame or praise, we need to understand why it's not working. Group dynamics and a bit of sociology can help.
>> 
>> Our Coordination Group is different from what some of us/you have come to consider as "normal". The technical bodies (IETF, IAB) have developed an efficient process where "rough consensus" is understood and accepted. But other components of the ICG have different habits, and also a different accountability mechanism: however attractive "rough" may be, it is insufficient. For example, the GAC has its own rules (a joint position can only be reached by unanimity), and the ALAC routinely conducts all its votes on a full-membership basis (each member has to say ay, nay, abstain, or be noted down as not having cast a vote).
>> 
>> So the challenge is this: is the "rough consensus" really adapted to all the needs of our group? With the experience gained collectively in London, and especially since then, I would recommend a dual approach:
>> 
>> A/ MATTERS REQUIRING ALL MEMBERS TO VOTE (typically, to be decided as soon as possible, with the exception of our Transition plan)
>>   - Chair structure and membership,
>>   - Charter of the ICG,
>>   - choice of Secretariat (ICANN or outside of ICANN, or a mixture of both),
>>   - choice of near-final drafts and approval of final draft of our Transition plan, before presentation to the NTIA.
>> 
>> B/ MATTERS WHERE OTHER FORMS OF DECISION-MAKING ARE ACCEPTABLE
>>   - Appraisal of specific community input, as a contribution to the ICG's recommended plan (e.g. ALAC should appraise input from its own community before submitting it to the whole ICG),
>>   - external relations and communications of the ICG (once the Chair structure has been chosen and populated, it may wish to ask Chair, or another of its members, to be the point of contact),
>>   - administrative & logistic matters, in conjunction with the chosen Secretariat (here too, delegation would be possible).
>> 
>> I'm prepared to provide a more detailed proposal for the above items.
>> 
>> Best regards,
>> Jean-Jacques.
>> 
>> 
>> 
>> ----- Mail original -----
>> De: "Kavouss Arasteh" <kavouss.arasteh at gmail.com>
>> À: "Patrik Fältström" <paf at frobbit.se>
>> Cc: "Coordination Group" <internal-cg at icann.org>
>> Envoyé: Lundi 11 Août 2014 10:40:08
>> Objet: Re: [Internal-cg] Consensus building process
>> 
>> 
>> 
>> 
>> Dear Wolf
>> Thank you very much for reply
>> My point is that if one or more ICG Mmember(s) is7are againszt the ruling of the Chir ,They could raise their issue and the matter must be settled by simple explanation or if not resolved by voting . I.E. CHAIR DOES NOT HAVE DECISION MAKING POWER ON HE OR HIS OWN WISHES RATHER TO TAKE INTO ACCOUNT VIEWS OF MEMBERS Regards KAVOUSS Regards
>> 
>> 
>> 
>> 
>> 
>> 2014-08-11 8:33 GMT+02:00 Patrik Fältström < paf at frobbit.se > :
>> 
>> 
>> 
>> 
>>> On 11 aug 2014, at 08:09, WUKnoben < wolf-ulrich.knoben at t-online.de > wrote:
>>> 
>>> The chair’s designation that consensus is reached is not her/his own decision rather than a wrap-up of extensive discussions. Of course this designation can be challenged by members. And this is what triggers your question about “If several participants in the ICG disagree with the designation given ...”. I’m open to any helpful suggestion on how we could procede in such a case.
>>> In the end consensus - as defined – has to be achieved.
>> 
>> Let me emphasize what you say here, which I strongly agree with.
>> 
>> We must deliver.
>> 
>> This implies we must be able to reach consensus.
>> 
>> The last couple of weeks discussions on various topics makes me a bit pessimistic on the ability for us to reach consensus, but I am optimistic, always optimistic, on peoples ability and interest in actually deliver.
>> 
>> Remember that the chair is calling on the consensus question, not the substance. That way the power of the chair is decreased to a minimum and process issues.
>> 
>> Patrik
>> 
>> 
>> 
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> 
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> 
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
>> 
>> _______________________________________________
>> Internal-cg mailing list
>> Internal-cg at icann.org
>> https://mm.icann.org/mailman/listinfo/internal-cg
> 



More information about the Internal-cg mailing list