Re: addendum to the gTLD MOU: phase 1: building a list

[Jim Dixon <jdd@matthew.uk1.vbc.net>]

On Fri, 19 Sep 1997, Amadeu Abril i Abril wrote:

>   1       the financial barriers to CORE membership should be low
> 
> This not a matter for gTLD-MoU, even not for CORE-MoU. 

Oh, we can put anything into the MOU that we like.  We can put "pi = 3"
in, if we like; I think that at one time the Tennessee legislature did
just this.  

There is a serious point here: you assume that there is a set of real
rules, unspoken ones, that everyone must follow.  I don't.  I assume that
we are starting with a clean slate.

>   2       CORE should begin operations with a large number of registrars,
>           hundreds if at all possible
> 
> Yeah, everyody being forced to be registrar.

Don't be silly.  There are 420 registrars in the UK, a country of 60 
million people.  So I would expect that recruiting several hundred members
for CORE would be easy.  

>   3       CORE should begin operations as soon as possible
> 
> There we agree. But your move could only lead to stop current selection process
> and start everything form scratch is quite inconsistent with this goal,as far as
> I can see.

I didn't say that everything should be started from scratch.  I think that
the process of getting CORE up and running should be accelerated.
 
>   4       CORE should be professionally managed
> 
> This is exactly the kind of  provision which inclusion in a
> basic,onstitutional-like doc as the gTLD-MoU is essential. 

The current plan is for CORE to be a committee.  The actual business
will be run by an external vendor.  This is dangerous.  For one thing,
this is what turned NSI into Frankenstein's monster.

> Jim, this is an OPERATIONAL issue that has no place in gTLD-MoU. Nor in any
> other doc. How do you enforce such a provision? What do you mean by
> "professional"?

Do you really need a definition?  CORE should hire a competent manager
and other staff; they should run the registry.
 
>   5       CORE should run the gTLD registry itself, the registry should
>           not be contracted out
> 
> Yes, and we will prevent General Motors  from buying or subcontracting outside
> its conglomerate.

? 
 
>   6       if at all possible, existing shared registry software should
>           be used in CORE's operations
> 
> "Existing shared registry"?? Now you sound funny. Are you against progress? Has

No.  I want CORE running as fast as possible.  The fastest way to get it
up and running is to use existing software and then improve it.

>   7       CORE should be strongly encouraged to use open standards
> 
> CORE should be encouraged to use reliable software. Use of often standards is
> indeed desirable, to the extent it  is possible and never to the expesnes of
> quality of service.

This is a list of the things that I would like to see.  If there are 
strong objections to open standards, we take it out of the list.
 
>   8       CORE should be strongly encouraged to use public domain software
> 
> Public domain db that large, powerful and reliable? Well, if they exist.... But
> again fitting this is a
> second order choice, after service quality.

Same reply.
 
>   9       CORE should be strongly encouraged to cooperate with other
>           registries
> 
> Oh, and if they don't want to cooperate, we send Perry, Bob Allisat and me to
> force them, nicht
> wahr? But wait, perhaps you have got a call from NSI or ES-NIC....

Your reply is odd.  There are other registries in the world (RIPE, 
the InterNIC, the other NICs, the national registries).  Everyone would
benefit if these adopted common formats for data exchange (the RIDE
proposal) and if there was common, public domain registry software in
use by all of them. 

The Internet runs on cooperation, open standards, etc.
 
>   10      a gTLD name server should put put on line as quickly as
>           possible, preferably by 15 October
> 
> Perfect, before the selection period ends. Or still better, before you convince
> us to redo the whole thing. Any idea where could we locate this server?

Well, all sorts of people would house it voluntarily.  

A name server that has nothing in it but test data could be a $500 Linux
box.  Any computer science student could bring it up.  This is not
rocket science.

But as I have said before, we don't know that NSI will point the root
name servers at a CORE gTLD server.  This is a fundamental point.  I 
would test it today if I were one of the people running IAHC/PAB/POC/CORE.

If NSI won't point at the CORE gTLD server, there is no reason for any
of this: the gTLD MOU, CORE, the entire process.

>   11      a good understanding of some aspect of Internet operations
>           should be prerequisite to having a vote on the POC (that is,
>           if you are going to tell people how do do things, you should
>           know what you are talking about)
> 
> Ooops. I DON'T vote on iPOC.. Add that some good understanding on what to put on
> a MoU should
> also be a prerequisite. The P of POC stands for policy, if you don't mind.
> Techcnical advice can
> be contracted outside. Policy sensiubility is more difficult to find.

"Technical advice can be contracted outside."  But the ability to 
judge whether that advice is valid cannot.

The primary purpose of the POC is to make policy decisions regarding
DNS.  You should not be making such decisions if you do not have 
some understanding of the practical, real-world effects of your 
decisions.

>   12      non-voting, advisory seats on the POC should be open to
>           representatives of various communities affected by the
>           Internet
> 
> Wholly agreed, but replace Internet by DNS. gTLD-MoU is not about governing
> Internet. It is adding a new DNS management model.

You can't have the Internet without DNS, not the Internet as we know it.

>   13      the ITU and WIPO should immediately lose any votes on the POC
> 
> I note you say votes, not presence. I agree.

I meant exactly what I said.  Advice from the ITU and WIPO is valuable and
should be heard.  But decisions should be made by those that represent the
Internet.
 
>   14      domain names are addresses
> 
> Yes, and the millions of people who may think otherwise should be reeducated .
> Period. If everybody could agree that they are ONLY addresses we would not be
> discussing all this.

You are right.  They should be educated.
 
>   15      the rights of trademark holders should be recognized, but only
>           in cases where there is a clear attempt by a domain name
>           holder to pass himself off as a trademark holder
> 
> I'd could devise more precise legal terms ;-) I notice you are not bothered
> about reverse hijacking, or other ways of pirating htat occur.

The words "piracy" and "hijacking" imply the existence of ownership 
rights.   I don't agree that such rights exist.

If Joe McDonald wants to set up mcdonalds.org and he is not pretending 
to be that hamburger shop, I see no reason at all why McDonalds should
be able to take the domain away from him.  If he tries to use the domain
name to deceive others, OK, he loses it.

>   16      the Internet community should recognize the contributions of
>           the members of the IAHC and the iPOC
> 
> And PAB. And registrars. And maillist contributors. And fierce opponents. But
> not as an addendum ot a MoU, Jim. Let's be serious [...]

I was just trying to be nice :-)

> Apart from building Nominet II and promoting some rather day-to-day operational
> issues as the
> Ground Norms for Internet Governance Evolution, do you have any other
> suggestion?

I will think of some more over the weekend.
 
> I translate my mail: I'm not saying that any of your proposals is nonsense. Most
> are not quite the contrary. I'm just pointing out that I CAN'T BELIEVE that the
> thousands of people you claim are wihing to enter the system but are prevented
> by the current content of gTLD-MoU would be much impressed by these amendments.
> And that many of them would simply read ridiculous in a MoU, no matter how sound
> they are.

We have no way into the system other than by signing the gTLD MOU, and that
is unacceptable in its present form.  So the first thing to do is to 
seize control of the PAB and amend the MOU so that it gives us enough
power to take the second step.  That is to take control of the POC.
Once there, I suspect that we could rewrite the ground rules and get 
control of CORE.

A simple and moderately ambitious plan :-) 

I am sure that somewhere in Geneva someone is laughing wildly.

--
Jim Dixon                                                 Managing Director
VBCnet GB Ltd                http://www.vbc.net        tel +44 117 929 1316
-----------------------------------hobbies---------------------------------
Member of Council                                                 President
Internet Services Providers Association                       EuroISPA EEIG
http://www.ispa.org.uk                              http://www.euroispa.org
tel +44 171 976 0679                                     tel +32 2 503 2265