[discuss] Thoughts welcome on proposed Netmundial submission

[John Curran]

On Mar 2, 2014, at 5:41 AM, Peter Dengate Thrush <barrister at chambers.gen.nz> wrote:

> To me, this substantially misses the major point, and appears confused about the current set up.
> 
> IANA functions do not have a separate entity - they are are a set of functions performed already by ICANN.
> That means they are already a function under the auspices of ICANN.

Agreed.

> THE issue in play is that ICANN performs these functions under a contract from the USG.

Or even more precisely - under contract from the USG and per agreement with the 
IAB/IETF (RFC 2860)

> The issue is that this contract provides the USG with an unequal advantage compared with other government, and all other IG actors.

Yes (even if only used with the purest of intentions, the very presence of the contract
results in a larger "seat at the table", and hence potential for advantage in discussions)

> Making progress on this issue means removing the contract, and the contractual relationship that it creates. The USG originally signaled in Clinton's time that it would complete a handover to the NewCo of the White Paper, once it was built. One avenue is therefor to return to the USG and say that ICANN has now become the trusted global body envisaged by the White Paper, and that transfer should now be done.

Also agreed that this is a potential route forward (following the original intent of the White
Paper towards handover at a future time)

> How that gets done is important - there should be no opportunity for US politicians to argue that the Administration "gave away the Internet".
> 
> Time in Singapore and Brazil should productively focus on the mechanisms by which this transition should occur.
> I have suggested several in the past - eg treating the IANA functions as "trust" property, with ICANN as Trustee. If necessary, the USG could retain a role settlor of the Trust, with the power to appoint trustees - not a huge distance from the current (contractual) situation, but a lot better in practice. ICANN would "own" IANA, the beneficiaries would be the Internet community. The terms of the trust would include that ICANN operate IANA under DNS directions frm ICANN, etc etc etc.
> 
> Or, the USG could "assign" the IANA functions, with a security ( like a mortgage) over them. The terms of the mortgage would be the similar to the obligations in the Trust deed above.
> 
> This would allow the USG to exercise a right of re-entry if the terms of the mortgage were breached.
> 
> These arrangements need to satisfy both masters - the global internet community, looking for a way to lessen the USG role and US politicians who see the Internet (wrongly) as a US fief.
> The solution should be very long term, so that there is stability.

It might be worthwhile for you to submit one or more papers outlining some of the potential
arrangements.

>> It is an interesting formulation of the problem statement...   At present, I would describe 
>> the IANA functions as "a set of tasks" rather than an "entity", and hence would instead
>> phrase the purpose of a roadmap as:
>> 
>> "This roadmap suggests that the IANA functions (which are necessary for the secure 
>> and proper functioning of the Internet) that are currently administered by ICANN per
>> USG contract should remain at ICANN and be performed instead under its auspices 
> 
> better
>> via the strengthening of accountability mechanisms to meet the global public interest.
> 
> but this is a non-sequitur.
> Improving the accountability mechanisms ( a most worthy goal, which needs to be built in to the DNA) wont keep the IANA functions at ICANN.
> 
> 1 - they are already at ICANN and there is no suggestion they should be moved, and 2. they are there under a contract, which is the root cause of the problem.
> 
> Nor will improving accountability keep them being performed under ICANN auspices, if the USG decides to re-let the contract, or take the functions in-house.

Apologies for lack of clarity - improving the accountability mechanisms may be necessary for 
the community to safely entrust the functions to ICANN in the absence of any direct USG/NTIA
oversight via the present contract.

The fact that the functions are presently performed at ICANN under some credible representation
of the will of the Internet community does not imply that will be the case in 5 or 10 years if there is
no particular reason for ICANN to do so.   The AoC has been cited as one such mechanism for
accountability (which could survive absent an specific IANA function contract), the question is 
whether that (plus ICANN's own legal structure of accountability) suffices.  It is also the case that 
the particulars of the handover (as you describe above) may provide sufficient assurance, but 
the point is that as a community we have enjoyed a certain level of stability which needs to be 
maintained.

>> The above criteria confuses me - are we referring to ICANN's DNS policy development role, 
>> or performance of the IANA functions?   The latter are technical tasks in registry administration
>> and the most important criteria would be that ICANN continue to implement all IANA registry 
>> functions in accordance with the respective policies (I guess one could further elaborate to 
>> point out that operating per respective policies means free from political or other interference,
>> but that really is secondary to making sure that the IANA follows IETF protocol, RIR IP, and 
>> ICANN DNS policies, both presently adopted and as revised in the future.)
>> 
> What is needed is cut and paste from the SOW with NTIA, so that these controls (worth and necessary)  which are there are captured are continued in the new ( non-contractual) instrument, plus a community debate about the appropriateness of new and other controls.

Agreed as well.

> Another topic that should be prominent on the agenda's in Singapore and Brazil.

Looks to be some busy times ahead...  :-)
/John

Disclaimer:  My views alone.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://1net-mail.1net.org/pipermail/discuss/attachments/20140302/c88f55f1/attachment-0001.html>