[ncdnhc-discuss] Internet is global=we need central planning

[Dave Crocker]

At 03:42 AM 5/2/2002 -0400, Michael Froomkin - U.Miami School of Law wrote:
>On Wed, 1 May 2002, Alejandro Pisanty - DGSCA y FQ, UNAM wrote:
> > The degree to which you seem to think that some functions can be
> > decentralized appears unrealistic to me. As a prime example, gTLD policy.
>
>**Why** exactly can't this be decentralized?

Michael, when someone calls for a change, they have a responsibility to 
provide details that show the efficacy of the change.  It is not reasonable 
to call for a change, provide no implementation detail, and then demand of 
others that they demonstrate why the change is not appropriate or feasible.

So if you are going to claim that it is appropriate to change an 
operational and administrative model that has been successful since the 
inception of the DNS, more than 15 years ago, the burden is on YOU to 
provide documentation concerning feasibility and efficacy.

It is no wonder that you are disinclined to provide such detail, given the 
absence of any real-world, comparable examples to demonstrate the 
feasibility and efficacy of your proposal.


>As far as I can tell, the only functions that require centralization are
>...
>3. setting in motion the processes by which other bodies will choose the
>gTLDs;

You need to specify in detail how those "other bodies" would achieve 
allocations that do not collide with each other and how this would be 
superior to the current model.  That specification needs to attend to 
disputes about allocations.


>1 is easy.
>
>2 becomes much easier if one accepts that ICANN should look only at the
>engineering issues.

Most things are easier if one ignores the hard parts.

Technical administration and operations is a hard problem on a global 
scale.  When you start attending to that reality, discussion about 
proposals will have some grounding in reality.


>   But after three years, and in the face of some
>fairly strong consensus that if there's a technical limit it is very high,

This is a good example of trivializing the topic.  The maximum number of 
TLDs that can be supported in the root servers is only one of several 
issues concerning limits.  That there is consensus among DNS experts that 
that physical limit is high -- and there is far less consensus about just 
how high -- is fine, but not nearly sufficient.

This topic has been explicated many times over the last few years.  Before 
asserting a simplistic truth about this, please familiarize yourself with 
the broader range of issues affecting DNS operational limits.


>I think the burden is quite clearly on those who would leave the limit a 7
>every three years.

There is no such group.  Please do not invent groups and positions to 
oppose.  It does not facilitate understanding or progress.


>And, if it needs saying, the idea the ICANN has security
>*responsibilities* is a total non-starter.

The idea that you can assert an absolute over a topic with which you 
demonstrate far too little understanding is an even better non-starter.

Again, the range of issues involving DNS security has been discussed at 
length. Please familiarize yourself with those discussions and try to 
provide specific, informed comment rather than simplistic, broad dicta.


>   The fact that it has appointed
>a committee -- the prime evidence for this responsibility noted in the
>staff report-- is evidence of nothing other than desires for grandure.

You really need to stop projecting psychological assessments of yourself 
onto others.

d/

----------
Dave Crocker <mailto:dave at tribalwise.com>
TribalWise, Inc. <http://www.tribalwise.com>
tel +1.408.246.8253; fax +1.408.850.1850