Re: [Internetgovtech] Transition to the web

[John C Klensin <[e-mail]>]

--On Saturday, July 12, 2014 11:15 -0700 Eric Brunner-Williams
<[e-mail]> wrote:

>...
>>    The second purpose is to pay the
>> relatively nominal cost of administration of the IANA.
> 
> I've made a cursory glance at the current budget DRC was kind
> enough to provide a link to, and $2.6m/yr appears to be the
> incumbent contractor's projected cost to carry out the IANA
> Function. That's 17 cents per "Community Anchor Institution"
> (schools, etc., with broadband), according to the USG [1] or a
> fractional penny per installed CPE within the United States
> [2], or about two quarters per square foot of data center --
> just in Loudoun Country (Northern Virginia) -- I'm going to
> guess that amortized over aggregated data center acreage for
> the rest of the United States this number approaches a penny
> per foot. Budget reading, math and industry stats error
> corrections appreciated.

Assuming that your calculations are correct (I suspect they are
close enough for analysis purposes).   The "17 cents per..."
style of analysis seems to me to be useful in only two sets or
scenarios:

(1) To claim that IANA is a public good that should be paid out
of the public coffers without worrying too much about the
amounts involved.  If one makes that argument, one must, I
think, be prepared to argue that IANA is more important (in
terms of priority access to the public coffers) than other
public goods, such a preventing starvation, educating small
children, sundry emergency services such as police and fire
departments, etc.  I'm disinclined to go there; YMMD.

(2) If one has taxing authority with regard to the sources whose
resource impact is being measured and intends to use it.  ICANN
clearly does not have that authority (independent of whether it
can assess fees on entities who do business with it, a different
matter) and a further separation of ICANN from the US Government
also further separates it from anyone with such authority.
Even if ICANN's (or IANA's) ties to the government were closer,
I don't think additional (or new) taxes, especially on
"Community Anchor Institutions" who are mostly nonprofit and
tax-exempt would have much chance in today's legislative
environment in the US.

All of that is obvious independent of the question of whether,
if IANA is going to be treated as a public good or the
equivalent thereof (as several of the discussion threads on this
list effectively suggest even if they don't use those words), it
isn't at all clear why the costs should be measured, assessed,
or supported out of the US alone. 

The latter takes us to another issue to which I know you are as
sensitive as I am.   If the argument for community support of
IANA (generally and the protocol registries in particular) is,
as others have suggested, because everyone using the Internet
benefits from those services, and that principal source of that
support is DNS name-related revenues, then it seems to me that
there is no excuse at all for the ccTLDs --at least the  ccTLDs
who are operated on other than an cost-recovery basis-- being
exempt from paying for those services.    

IMO and for all sorts of reasons, it is not a good idea to try
to insist on that, but that suggests to me that the analyses of
the model and motivations need more thought.

   best,
   john