[discuss] governments and rule of law (was: Possible approaches to solving...)

Thu Feb 27 22:01:41 UTC 2014

On Feb 27, 2014, at 4:00 PM, Alejandro Pisanty <apisanty at gmail.com> wrote:
> so we are not that far:
> 
> 1. Boards must be expected to exert discretion - the least often the better, the better done the better. (Plus: the reasons Boards and other collective decison-making bodies exist is that given the impossibility of fully objective decisions, a vector sum of subjectivities is the least-bad alternative.)

Alejandro -

   To be clear, the issue is not the Board determining what is in best interests to the mission 
   of the organization, it is that such a determination should generally consider upholding the 
   existing commitment by ICANN to bottom-up community-driven processes and outputs to
   be the result which is generally "in the best interests" of ICANN and the ICANN community
   (unless there is a clear evidence of failure of those processes)   The default of supporting 
   the status quo policy until there is clear and well-supported consensus for change should 
   be paramount; even deadlock can be useful, as it supports stability and encourages parties 
   to truly listen to one another and seek common ground.

> 2. In a single-purpose, relatively homogeneous organization like an RIR, things may be simpler.

  I'm sorry, your characterization of RIRs as "single-purpose, relatively homogeneous"  
  organizations as contrasted to ICANN is faulty...   We have ISPs, end-users, businesses,
  civil society, law enforcement, academia, government ministries, etc. all attending, and have 
  discussions of many of the same issues that are dealt with in ICANN's DNS policy world 
  (privacy vs accuracy vs LEA access for Whois, issues of identifier equity and fairness
   for developing economies, property rights vs community policy vs sovereignty, etc.)

  The main difference is that such discussions take place on the main floor of our meetings,
  where the points expressed are weighed by all participating based on their merits, rather
  than being fully formed positions advocated between "constituencies", i.e. we utilize the
  same approach as the IETF (where there are also a variety of attendees.)  i.e.  those 
  parts of the I* system that don't choose to separate the community into constituencies 
  will definitely will appear be more homogeneous from the outside.  

  ICANN's mission was actually supposed to be _coordination_ of policy development via
  distinct supporting organizations;  the increased complexity that it suffers is the result of
  bringing that inside ICANN for DNS and the related creation of the various constituencies.
  (There are also some pretty significant accountability aspects, since it is very difficult for
  a Board to provide impartial oversight to policy development when also deciding policy,
  but that's not complexity per se as much as conflict of role.)

> Where we may differ is whether ICANN becoming a membership organization is the way to go. Structuring the membership so as to achieve reasonable balance, and checks-and-balances mechanisms, between disparate entities is likely to land very close to where ICANN is now, or a foreseeable step of evolution. Reducing the extortion power of non-committed entities may require quite elaborate rules, tons of work, and very little ROI for the effort as well - in that you may not end up very far from where we are now.
> 
> I don't think anyone is fundamentally opposed to see an exploration of these redesign options but only few are actually going to give it a go. 

   Start small:  provide a removal mechanism which can be invoked by a supermajority
   of the supporting organizations/constituencies.  Membership does not have to go to
   each and every Internet user (although it certainly could); there are many membership
   structures which recognize other organizations as their class of formal member, and 
   they do not require "elaborate rules" or "tons of work."  Trying to make the perfect model
   on day one should prevent initial high-ROI steps to address the problem.  Some would
   even argue that a formal membership isn't required, so long as there is some structure
   for Board accountability even if solely to internal constituencies and with a very high bar.

> In the meantime a non-membership organization is what we have and in many ways it has far more features than bugs. 

  You should consider that the current lack of formal accountability of the Board to any defined
  membership may be seen a fairly serious "bug" when it comes to ICANN and its suitability to 
  operate in the absence of some formal oversight mechanism.   Whether all of the "features" 
  you allude to make that an overall worthwhile tradeoff is for the community to consider...

> The only value I see in having continued this discussion on the 1Net list is that it may yield usable lessons learned for work further afield, if anyone, in 20 years, reads this again. 

  Wow; I would have thought that we'd be having quite a few discussions about accountability 
  over the next several months, but that could just be my misreading of the present situation.

Thanks!
/John

Disclaimer:  My views alone.