[arin-ppml] Can a personal property approach ever transition into multi-stakeholder, private sector led, bottom-up policy development model?

David Conrad drc at virtualized.org
Fri Apr 29 11:58:39 EDT 2011


On Apr 28, 2011, at 10:05 AM, John Curran wrote:
> On Apr 28, 2011, at 2:31 PM, David Conrad wrote:
>> The reality (at least as I see it) is that at least some legacy holders believe (if they consider the question at all) that the IP addresses in their inventory are their asset to do with as they please.
> Let's explore that statement a little further to see if we can get
> some clarity on underlying issues...


> 1) It's the position of many stakeholders that we should operate
>  in a model of multi-stakeholder, private sector led, bottom-up 
>  policy development model of technical coordination for these
>  identifies that acts for the benefit of global Internet users.


> 2) Such policy development model should not only allow open 
>  participation, but ideally should be based on the voluntary
>  contractual agreement of participants for purposes of 
>  implementation of the developed policies.  This also means
>  that no governmental or regulatory mandates are necessary
>  for the system to operate effectively.

I disagree with "ideally should be based on the voluntary contractual agreement of participants".  Contracts are a crutch, while convenient are not a requirement and would be, I suspect, an impediment.  Ideally, people should work together cooperatively without the threat of contract lawyers hovering over them.

> 3) Presume an hypothetical organization (JKL) that does this for 
>  all IP address holders globally, and also provides for big
>  picture overall policy goals like periodic recompetition,
>  integrity protections, transparency, accountability, etc.


> Once that's all up and running just fine, how do the existing 
> legacy address holders transition into the system?  

By creating an environment in which it is in their interests to do so.

> It is clear
> that has to happen if you want to be able to evolve and set 
> effective policies systemwide,

Is it? It is clear it would have to happen if you want to impose policies without consent.

> A government mandate or ruling to that effect is suboptimal, since 
> that's contrary to the principles that we are building something 
> that doesn't require such to operate.

And would likely be a bit challenging in the context of "globally".

> Can we actually have *any* multi-stakeholder, private sector led, 
> bottom-up policy development system for technical coordination 
> of identifiers at the same time as we have parties presently that: 
> 1) have no existing contractual requirements to participate in the 
> system, but 2) presently hold number resources in their belief as
> personal property?


There is already a limited example in the DNS space: IANA registry services to the ccTLD community (and note I'm speaking here specifically of IANA root management services, not ICANN, so ICANN bashers can chill). 

IANA policies for technical coordination of those identifiers are multi-stakeholder, bottom-up developed, and community-led (there are public sector folks but they are peers with private sector folks in the context of IANA policy definition).  With a tiny number of (voluntary) exceptions, there are no contractual relationships and there are definitely no contractual requirements (something that had to be learned by the folks who operated IANA early on) to participate in the system and trust me when I say that ccTLD admins view their resources as their property.  

(Yes, I'm well aware the IANA functions are provided by ICANN under contract to the USG, however the point is that IANA provides services to ccTLD folks without contract. ICANN would provide IANA root management services to the ccTLDs even if the USG wasn't involved.)

Admittedly, in many ways this can be sub-optimal (e.g., changing policy (even trivial and obvious policy) can take a long time, there are ccTLDs that are obviously doing stupid, broken, or even evil things and IANA staff can't stop them unless it directly impacts Internet security/stability, and funding for those services is (arguably) subsidized by external parties), however in reality, it is the _only_ way the relationship between IANA and the ccTLDs was workable.  

I don't know if this sort of model can scale to the size necessary to deal with address space.  I'm sure there are other models in other fields.  The ccTLD/IANA one is just one I'm quite familiar with.

However, I suspect what has to happen before there can be any sort of transition is "the ARIN community" has to understand that ARIN is a coordinative body *NOT* a regulator, just as ICANN (as the IANA root management function operator) had to understand that it was not the regulator of the ccTLD community, rather its role is to facilitate ccTLD admins coming to consensus. The real question is whether or not "the ARIN community" can do this before being overtaken by events. 


More information about the ARIN-PPML mailing list