Re: PAB The Green Paper and competing registries

John Charles Broomfield (jbroom@outremer.com)
Fri, 27 Feb 1998 10:46:02 -0400


Jim Dixon wrote:
> On Fri, 27 Feb 1998, Patrik =?iso-8859-1?Q?F=E4ltstr=F6m?= wrote:
(...)
> > Well, it all depends on what you think you will compete with, and how, you
> > can see what you will fulfil with competition between registries. It is too
> > easy to just say that you want competition between registries.
>
> This is not what I want; it what the chair of the meeting, the DTI
> representative, said was an absolute requirement.
>
> If pressed, he might waffle on this. But it would be sensible to
> assume that he meant what he said and that this is the position of
> the UK government, who currently hold the European Union presidency.
>
> I believe that you would get much the same response from DG IV, the
> European Commission's department dealing with competition.

I understand completely that Jim wants to get a solution and that this
solution, no matter how much we dislike it HAS to be acceptable to the
politicians. It's too late now to just tell all politicians "sod off".
Now, unless I've got things completely wrong, "absolute requirements"
from
politicians are not absolute at all.
Obviosuly if the politician is asked "do you prefer a monopoly registry
or would you prefer competing registries?" he'd probably answer the
latter.
A bit of coaching for him might be a good idea

(...)
> > Because of that, it can only be one registry per TLD.
>
> That doesn't make any sense. The CORE shared registry was designed to
> host many TLDs.

Patrick says one registry per TLD, not one TLD per registry. Each TLD
has only
one registry. It doesn't conflict with the design of CORE.

> Similarly on price. I think that you are being idealistic in claiming
> that prices will inevitably fall. If you look at the monopoly nTLD
> registries across Europe, you will find many different pricing policies.
> Some registries are run very well and have low prices to registrars;
> some are run very badly, have bad policies, and have high prices.
> Where CORE sits on the spectrum will be largely determined by chance.
> In the longer run, I would certainly expect a slow upward drift in
> internal costs. Bureaucracies are like that.

I don't think that the comparison is good. Currently only the nominet
model
would be a comparable example. We're talking about shared registries
that
should be run on a non-profit basis, where it is in the best interest of
the registrars for that service to be good and cheap. (I'm not sure, but
maybe the french model with the changes introduced Jan-1st this year is
going
in that direction, also it seems that the Canadian registry is ALSO
going
in that direction).
So pointing at what the OTHER nTLDs are doing (when they are mostly run
as
the equivalent to private owned) doesn't really seem to make much sense
to
me.
In any case, for CORE to have prices slowly increasing would mean that
the
registrars are happy with it, which would seem ludicrous to me...

Yours, John Broomfield.