Re: [wg-c] Re: IP/TM Concerns & New GTLDs

[Kent Crispin <kent@songbird.com>]

On Sun, Aug 01, 1999 at 11:17:43PM -0400, Milton Mueller wrote:
> Kent:
> 100 gTLDs does not necessarily mean 100 new registries, and it is the
> number of registries, not the names as such, that matters most from a
> regulatory standpoint.

This matter is under dispute, of course.  If the single registry 
were run as a non-profit cooperative, as Nominet is, then the 
competition essentially gets down to the registrar level.  The 
single registry is under the watchful eye of anti-trust authorities; 
it is a non-profit; etc.  The difficulty of regulation becomes 
greater as you have more registries, given that all registries are 
monopolies.  [This matter is also under dispute, and basically 
revolves around how serious a problem one thinks "lock-in" is.]

> Perhaps we can start to come to an agreement if I clarify how I approach
> the problem. I approach it by starting with the question: what do we need
> to do to have effective competition in the gTLD market?

It's easy -- shared registries, many registrars.  This model is
proven to work, in the case of Nominet, and the German registry. 

> I think we can all agree that authorizing one new gTLD registry is not a
> solution to the current situation. We would simply create a new NSI, one
> that runs dot firm or dot web instead of com, net and org.

I don't agree with your characterization of the end result -- there 
are many complications -- but I agree that it would be nice if there 
were multiple registries/registry operators.  What we disagree about 
is how quickly they need to be brought on line, and the business 
model for them.

> And it's no good to say that you're only going to give it that monopoly
> for a year or two. The advantage such a registry would have over its
> rivals, given the pent up demand, would be significant.

Only if you assume that the registry operator "owns" the TLD.  If it
doesn't, and it is run as a non-profit, then TLDs can be re-assigned
to new registries any time.  Recall that requiring the registry to be 
non-profit does not require the registry operator to be non-profit 
-- CORE is a non-profit, Emergent is a for profit.

> So you must have more than one new registry. How many should you have?
> There's a fairly well developed body of economic literature that says you
> need at least 5-7 competitors. Three competitors can too easily become a
> price-led oligopoly. It is just too easy for them to coordinate their
> pricing and strategies and avoid real competition. So it is better to
> push toward 7.

Sure.  Push toward 7.  Push toward a hundred.  That doesn't imply
that you need to *start* with 7. 

> Then there's the fact that each registry probably needs to service more
> than one TLD to be viable, to realize economies of scale. Let's face it,
> some names are going to be more popular than others. Each registry will
> need a repertoire of TLD names. Let's say, at a minimum, each will need
> three names.

Nominet is doing great with one name.  

It's important to realize that there are essentially no economies of
scale to worry about.  I could run a significant registry from my pc
at home.  This is technical fact.  It isn't widely known, but one of
the .com/.net/.org TLD servers actually is just a PC -- true, it's a
multiheaded Pentium server with a Gig of memory and lots of fast disk
and really good network connectivity.  But it's a PC.  And that's the
DNS server, which gets ~2000 hits/sec.  The registry database, as has
been mentioned before, gets orders of magnitude less traffic than the
dns server. 

The registry part of the registration business (as opposed to the
registrar part) is *insignificant*, relative to the registrar
component.  There are over a thousand Nominet registrars feeding a
single registry. 

> Five to seven registries operating three TLDs, gives you 15-21 as the
> baseline *minimum* that can be introduced if you want to have a
> responsive, competitive environment. We just can't talk about any number
> smaller than 15, and I'm not comfortable unless we're talking about at
> least 20.

You are vastly overestimating the value of competition at the
registry level.  As has been pointed out many times, a registry is
analogous to the small company that manages the 800 number database
-- a complete back-office operation, using known database technology,
with little of significance to be gained through competition.  There
simply isn't much social utility for competition at that level.  [The
development of shared registry software makes it somewhat more
complicated, it is true.  But it still is not a big deal.]

Even so, I think that there should be competition at that level, at
least in the long run.  But it is not a high priority.  On the other
hand, competition at the registrar level is very important.

> And that doesn't even take into account the many demands for non-English
> gTLDs, many extant claims for chartered gTLDs, the possibility of
> different business models (e.g., what if some of the registries want to
> be proprietary?).

Chartered TLDs are an interesting question, and in my opinion a 
great deal more consideration needs to be given to them.

> All of those questions can be resolved more easily if we have more TLDs
> to assign.
> 
> The smaller the number of gTLDs, the larger, more divisive and more
> unfair will be the political process of assigning them.

If the TLDs are proprietary, that is.  If we fix our sights firmly on 
the notion that gTLDs are *not* proprietary, then most of these 
issues disappear.  Almost all the real divisiveness centers on the 
notion of "ownership" of TLDs.

> The smaller the
> number, the harder it is to conform to good economic policy and the worse
> the results are likely to be from the consumer standpoint.
> Call it Mueller's law: there is an inverse relationship between the
> number of TLDs we have to work with and the prospects of an economically
> and politically beneficial result.
> 
> If you disagree with this analysis, and I am quite sure that you will,
> please explain how you propose to deal with the competitive implications
> of a small number, and how you propose to deal with the equity issues of
> assigning monopoly privileges to one firm.

I am not in favor of monopoly privileges for *any* firm -- go back
and read what I proposed: *three* non-profit registries to start
with, with more to come as more TLDs are added.  I also allow
for-profit registry operators, assigned to registries through
competitive bid.

To put it in concrete terms, what I would like to see is multiple
Nominet-style registries, with thousands of registrars able to
register names in any of them.  This is a proven model; it is working
so much better than the current NSI registry that there is really no
comparison. 

-- 
Kent Crispin                               "Do good, and you'll be
kent@songbird.com                           lonesome." -- Mark Twain