Re: Reasoned debate requires reasonable statements

[Kent Crispin <kent@songbird.com>]

On Mon, Jul 28, 1997 at 09:32:45AM -0400, Paul Svensson wrote:
> 
> IAHC/(i)POC/etc have done what they could about (3), ie giving good
> recommendations, and they should be commended for that.  They have
> actually stood up to (4), proposing an environment where competition
> could protect against overly large gains like NSI's. They've set up
> a structure (CORE/POC) that could take over the oversight responsibilities
> from NSF, to address (2), but ther don't seem to have any transition plan.
> They are obviously hoping that (1) will go away by itself at the end of
> the current cooperative agreement between NSF and NSI.

The departure of NSF from the scene does leave a vacuum, but I think 
that NSF abdicated any real oversight some time ago, so we have been 
existing in rather thin air for some time (to stretch the metaphor just 
a bit.)  Arguably, if NSF had maintained a strong presence the  DNS
registry wars would never have started.

The primary power NSF really brought to the scene is limitation of 
legal liability -- as a government agency they are relatively immune 
from suit.  Any group that steps in is considerably weaker in this 
regard.  Certainly the IANA is not immune to suit, nor are the root 
operators, nor are Jon Postel or Paul Vixie or any of a host of other 
players.  [Whether such suits would win is a totally different 
question, of course.]

In the absence of any other authority, the IANA has retained de facto 
control of the root zone, Paul Vixie/ISC has distributed the de 
facto root.cache, and the root operators use IANA's root zone.  There 
is no formal basis for any of it.

IANA signed the MoU.  ISC signed the MoU.  I don't know how many of 
the current root operators may have signed.

The IAHC Open Shared Registry plan does not seriously address the
issue of new TLDs (other than the proposed 7), or higher level
oversight of the root zone, but the key de facto authorities have
signed the MoU because they agree with the general principles outlined
in it -- the general principles are the great strength of the MoU, in 
fact -- almost everybody has something or other they don't like about 
the details.

As far as oversight, then, the MoU provides a framework, but no 
content.  However, the MoU also provides for self-modification and 
development, through a public democratic process, and the fundamental 
thing one is agreeing to when signing is a committment to the general 
principles and the process.

The most important thing to keep in mind, IMO, is that the MoU really 
only represents a *beginning*.  As I have repeatedly said, I don't 
speak for iPOC or PAB, but I would bet that every single member of 
both organizations are acutely aware that it is just a beginning.

[...]

> IAHC/(i)POC/etc have also erected a big complicated structure of conflict
> resolution panels, something that clearly is a solution in lack of a
> problem -- it doesn't address any of the problems, and it's not at all
> clear what thay would be good for, as there's no way to enforce their
> use.  It should be considered a separate, volontary effort, and not be
> connected to whatever changes are being done to the domain name system.

Obviously no one can say for sure how the ACPs will work in practice,
at this point.  If they don't work out, they *can* be dropped -- the 
MoU certainly is flexible enough to go to other solutions if they 
don't work.

> Both the IAHC and the other proposals seem attack the problems from the
> wrong end, probably because they see the top ones too difficult to handle
> directly, hoping their other efforts will be enough.

This is a matter of practicality, I believe.  

>  However, considering
> NSI's strong will and ability to push for status quo, I do not think that
> such hope has much base; I think a direct more approach would be needed.

Personally, I think NSI will sign the MoU within 2 years.

But you are correct that the Open Shared Registry plan -- I have taken
to calling it that, to see how I like the name -- has not dealt with
the issue of the NSI/NSF transition problem.  It is outside their 
charter, and, in fact, they have no de facto authority to put forth 
such a plan.

> Therefore, I would like to suggest the following plan:
> 
> 	1)	Replace NSF with someone willing and able to do the job.
> 		As it would take a while to set up a capable international
> 		body, let the US government handle it temporarily (sortof
> 		as proposed by NSI).  This body would take over NSF's
> 		part of the cooperative agreement, as soon as possible
> 		(before April -98 ??)  The international body to take over
> 		permanently could be POC/CORE (minus the ACP scaffolding),
> 		if/when they can get the governmental support they need
> 		to stand up to NSI.
> 
> 	2)	Let this NSF replacement negotiate with NSI about changing
> 		the policy, and make clear that the cooperative agreemet
> 		will not be prolonged as stands, and write "the letter" to
> 		make NSI give out their database.

You can suggest these; but it is like suggesting a plan for contintal 
drift.  :-)

> 	3)	Arrange the technical and operational means to share
> 		registration in the existing gTLDs, as I believe iPOC
> 		ar currently working on, and invite registrars to join.
> 		Do not limit who may be a registrar, in any way.

Restrictions on registrars are a short term aspect of the OSR plan.

> 		Do not require registrars to sign any political memoranda.

Not sure what you mean here.  Nobody has ever suggested that 
registrars sign political memoranda.

> 		Do not disqualify registrars for taking part in experimental
> 		registries.

The OSR plan does not disqualify anyone from doing this.  It only 
bars grandfathered registrations.  Any registrar who insists on the 
grandfathered registrations self-selects themselves out of the picture.

> 		Any unused part of any fees levied from registrars should be
> 		refunded to the registrars.

In practice this is not a problem, I believe -- the registrars have 
to come up with the money to get the system running; the fees will be 
totally used up by that.


> 		Do not build any conflict resolution system, just one simple
> 		rule:  first come first served, unless ordered otherwise by
> 		a competent court of law.  People who want arbitration outside
> 		of court can seek such themselves.

No need for this -- the conflict resolution system is defined, you 
don't have to use it; it doesn't conflict with court cases

> 
> 	4)	When this system is in place, and registration of
> 		.com/.net/.org is shared, it's time to consider creating
> 		new gTLDs, not before.

It's too late for this.  The odds of the OSR names being put in the 
roots are very high at this point...


-- 
Kent Crispin				"No reason to get excited",
kent@songbird.com			the thief he kindly spoke...
PGP fingerprint:   B1 8B 72 ED 55 21 5E 44  61 F4 58 0F 72 10 65 55
http://songbird.com/kent/pgp_key.html