[CCWG-ACCT] Governing vs co-ordinating (was Re: [community-finance] IANA Stewardship Transition - Project Expenses - FY16 Q3 update)

[Andrew Sullivan]

On Sun, Aug 14, 2016 at 10:26:56PM +0530, parminder wrote:
> 
> To say that it is up to anyone not to use the ICANN's DNS services today
> has more or less a similar meaning (except that there are even fewer
> ways to go to a place of its non-application).

Well, wait a minute.  There are _three_ ways that anyone can be
implicated in ICANN's DNS policies, and I think you're just collapsing
them.

One of them is to use the DNS at all -- to look anything up in it.
That is affected by ICANN policies only to the extent that gTLD
operations somehow affect one.  It would be interesting to me to
understand what issue you think there is here in the ways that ICANN
DNS policies might affect those performing lookups.  (Note that there
is a way in which one also interacts with ICANN in this, and that is
if one looks anything up in L-root.  One can solve that oneself,
however, with a different root.hints file.)

The second of them is to attempt to get a TLD of some kind.  To get a
gTLD, one has to go through an application when such applications are
being accepted, and pay a lot of money.  To get a ccTLD, one has to be
a country.  This is a clear example where ICANN policies are
restrictive of individuals, but that is the nature of being the policy
authority over a zone (in this case, the root zone).

The third of them is to attempt to register a name in a gTLD.  A
number of ICANN consensus policies do indeed apply to gTLD registries
and to accredited registrars in those gTLDs, and therefore this is a
way in which some ICANN policies might affect what people can do.  But
first of all, it is very far from obvious to me that the restrictions
in some of the gTLDs (the most open, like com, have very few) are
particularly restrictive at all, and are certainly rather less
restrictive than the sort of thing we see in actual countries whence
one might wish to flee.  Secondly, it is not like publishing a name
beneath a gTLD is the only way to publish online -- ccTLDs are not
subject to ICANN consensus policies; and ICANN policies anyway do not
apply recursively down the tree (making that clear was very important
to some of us in the discussions around the mission), so publishing at
another level of the DNS remains an option. 

> 'network' isnt either. And so one is, nilly willy, subject to all the
> policies that apply to the 'network', and ICANN policies is a part of
> it.

But it's not one network.  It's a network of networks with voluntary
inter-routing of packets without prior contractual terms.  This is
quite different than the relationship a citizen has to a state.
Waving that away is rhetorically convenient, I suppose, but it's not justified.

> I dont have to go into detailed descriptions of how ICANNs policies
> dictate online behaviour of all of us, and how many have, at their cost,
> found out the coercive power of these policies on their online presence/
> behaviour. This has been much discussed, including on this list. But if
> you are not convinced, we can have that particular discussion.

Yes, please.  Because I think you _do_ need to go into such detailed
descriptions in order to justify such claims as "dictate online
behaviour" and "ICANN is a regulator and it taxes".  Those are strong
claims, and I don't think it's reasonable to attempt to found them on
rather flabby ones about how this is all "much discussed".  I've
certainly heard such claims.  I've rarely seen them well justified and
I've too often seen them turn out to depend on a deep misunderstanding
of how the Internet actually works.

I'm not actually sure where this discussion belongs, however, since
I'm not sure this list is the place unless this has ramifications for WS2.

Best regards,

A

-- 
Andrew Sullivan
ajs at anvilwalrusden.com