[CCWG-ACCT] Definitions and the tussle (was Re: Fwd: [CCWG-Advisors] question regarding Global Public Interest)

Eric Brunner-Williams ebw at abenaki.wabanaki.net
Mon Dec 28 20:31:58 UTC 2015


Well, lets start with the allocation of a scarce resource -- ipv4 
addresses. Does the Corporation have an interest in the allocation being 
(a) congenial with routing, and possibly conservative as well (a subject 
of serious discussion on an RIR's policy mailing list), and (b) not 
captured by a single, or several, allocatee(s)?

Clearly there is a broadly held interest that routing work, and address 
exhaustion delayed as long as possible, and the distribution of 
allocations be somewhat uniform, reflecting shared goals of DARPA, the 
conversion from classful to classless allocation, and of course, Jon's 
farming out regionally the addressing component of his work at ISI, and 
a wicked large number of beneficiaries of these efforts to ensure 
routing, conservation, and at regional distribution.

We have come some way from the point in time when MIT campus held more 
allocated v4 addresses than all of the access providers in the PRC 
combined. The design of v6 allows at least one address per human being, 
a property absent in the v4 design.

Incorporating my note of the 25th, the Corporation Board has, over its 
nearly two decades of existence, observed that a public interest exists 
in access to numeric endpoint identifiers, and in access to mnemonic 
endpoint identifiers, unrestricted by region or language, and to some 
degree, only slightly restricted by access to capital, where packetized 
data communication is supported by communications infrastructure. This 
Corporation observation of public interests in access to endpoint 
identifiers is indistinguishable from the allocation behavior of the 
prior parties exercising "technical coordination", and so continuous, 
and likely to remain so in the foreseeable future.

The suggestion that finding a public interest is an exercise in 
sophistry would of necessity apply to the current, and prior, 
Corporation Boards, and those responsible for technical coordination of 
endpoint identifiers prior to November, 1998, specifically any 
representations that their acts to make numbers or names accessible to 
later adopters were in a public interest.

Eric Brunner-Williams
Eugene, Oregon

On 12/27/15 10:58 AM, Andrew Sullivan wrote:
> Hi,
>
> I'm sort of loathe to dive into this discussion, but I think there's a
> useful thread in here that is worth tugging on so that we can see the
> quality of the weave.
>
> My biggest worry about the phrase "the global public interest" is not
> the meaning of "global", "public", or "interest", but "the".  By
> claiming that something is or is not in _the_ global public interest,
> the definite article implies that there is such an interest (or maybe,
> such a public); that there is exactly one; and, perhaps most
> interesting, that one knows what that is.  Even if I were to grant (I
> do not, but let's say for the sake of argument) that there is a fact
> of the matter about the the interest of the global public, I cannot
> imagine how one would test a claim that something is or is not in said
> interest.
>
> The quest to come up with a definition of "the global public
> interest", therefore, is an attempt to create such a test; but it's
> really a dodge in a Wittgenstinean language-game.  Were we to unpack
> any such definition that was even widely acceptable, we'd discover
> either that some interest (or public) would be left out, or else that
> some conflict inherent in the definition would be obscured.  For the
> basic problem is that you cannot define "the global public interest"
> in a way that is all of universally acceptable, useful for the
> purposes of making tough decisions, and true.  Even apparently simple
> and obvious cases -- "It is in the global public interest for war to
> end" -- turn out to be troublesome.  For example, people fighting a
> current war are presumably doing it for some other end, so they'd only
> agree to that example statement with the implicit premise, "as long as
> my desired outcome is assured."
>
> A definition of "the global public interest" will be ever more
> troublesome the clearer it tries to be, because the list of specifics
> will start to be long.  I think our experience in working on the
> mission statement is mighty instructive, and it is at least scoped
> merely to the parts of the Internet ICANN directly touches -- whatever
> we think those are.
>
> As a consequence, I think a claim that _x_ is [not] in "the global
> public interest" is really just a way of saying, "I [don't] think _x_
> should happen."  Such a claim is part of a tussle, like the "Tussle in
> Cyberspace" described by Clark, Wroclawski, Sollins, and Braden (see
> http://dl.acm.org/citation.cfm?id=1074049).  It's a nice rhetorical
> move to claim that you can define the tussle away, but you can't (at
> least, not legitimately).  I think we should be honest with ourselves
> that such definitional efforts will create wheels that do no work.
>
> Best regards,
>
> A
>



More information about the Accountability-Cross-Community mailing list