Re: [wg-c] GTLD Straw poll

[Kent Crispin <kent@songbird.com>]

[A lot of what I say below has to do with the lawsuit-happy US culture, 
and would not necessarily apply in other jurisdictions...]

On Wed, Aug 18, 1999 at 12:21:06AM -0700, William X. Walsh wrote:
> Kent, your response below is so transparent, it needs no refutation,         > it refutes itself.

Perhaps it was a bit subtle.  Thanks for the example of the 
pharmaceutical industry -- I will use it below...

[...]

> Hmm, looks to me like we are seeing co-operation of a reasonable
> contract.  ALL of your concerns above are addressable.

Perhaps.  However, you have not addressed them.

> If contracts were such a bad instrument of enforcing restrictions in
> business relationships, then the corporate world is in big trouble.

This is a somewhat unusual "business relationship".  

Let me establish one principle I don't think anyone will disagree
with: practicality is a constraint on contracts, no matter how well written 
they are.  

This is because, fundamentally, the enforcement power for contracts
comes from lawsuits, and practical factors have great weight in 
determining the potential success of a lawsuit.  The result is not 
guaranteed, because the result is partially a contest -- this was so 
eloquently expressed by Tony Rutkowski in a recent message:

  "So the issue for the participants here - particularly the CORE and
  gTLD-MoU signatory participants is - do you feel lucky?"

We know, for example, that some lawsuits, even lawsuits with very
high moral or legal qualities, are not worth pursuing, because the
cost of litigation outweighs the value being litigated. 

Sometimes this effect is caused by the fact that the opponent is a 
large company with enormous legal resources.  That is, a large 
company can drive the cost of litigation very high, and can use this 
tactic to eliminate a small opponent.  

Moreover, there is no question that large companies with deep 
pockets can simply starve out opponents in legal matters -- examples 
are legion:  Tobacco companies have faced many lawsuits over the 
years, the famous anti-trust cases that have dragged on for years.  
I believe that in their own small way, NSI is trying to do this with 
ICANN -- we just haven't reached the lawsuit stage yet.

Even in cases where a lawsuit is prosecuted successfully, there is 
still the matter of enforcing the judgement.  There are many ways a 
wily opponent can avoid judgement.  (I am reminded of a time when I 
won a judgement in small claims court against a contractor...)

To repeat -- no matter how well written a contract, the enforcement 
mechanism that supports contracts has a number of practical 
limitations.  It is in the realm of such practical limitations that 
I don't think that contracts alone are an effective mechanism for 
ICANN.

I might add that there is also a strong political component to 
ICANN's situation that further complicates the situation.  This is 
obvious in the NSI case, but it is also going on in the ccTLD area, 
and we can expect ICANN to be susceptible to strong political 
pressures for the indefinite future.

> Jurisdiction is not an issue, and you know this.  I suggest you talk
> to a good contract attorney, Kent. None of your points above are
> salient.

Nice argument.  

> As to NSI's contract, well that "contract" (note: its not really a
> contract) is a poorly written instrument without many specifics.  The
> times have changed from when it was originally written, and we have
> learned a lot from that mistake.

Nonetheless, the relationship between ICANN and the registries is 
unusual, in a number of ways, and there is plenty of opportunity for 
contracts to miss something that later turns out to be a crucial 
problem.  NSF didn't set out to write a bad contract...at least I 
don't think they did.

> > It is far better if the enforcement is not necessary -- it is far
> > better if you establish structural constraints in the system that 
> > prevent the abuses in the first place.
> 
> Contraints must be justified.  You haven't shown why contractual
> obligations are not sufficient.  You support what is really a
> Draconian measure that places unnecessary restrictions without basis.

I will try to motivate that, below...

> The abuses can be prevented by making the contract strong enough.

Yes, they can, but then the contracts themselves become an issue. 
Furthermore, if the contracts are uniform, and strong enough, then
they effectively just implement the "structural" controls I was
describing.  For example, the contracts could require that the
registry run as a non-profit....

[...]

> Even the pharmaceutical industry, which is MUCH more vital and much
> more important than these petty domain name squabbles, have more
> looser restrictions than what you propose.  And talk about MONOPOLY!
> A company coming up with a new vital drug has a total and complete
> monopoly over that drug.
> 
> But adequate safeguards are in place, the minimum required to address
> valid concerns, without Draconian measures.
> 
> If it works for such a huge industry, where the stakes are literally
> life and death, and the opportunities for exploit FAR greater (we are
> talking BILLIONS of dollars) I see no reason the simple approach of
> minimal regulation can't work to address almost all of the concerns
> that the various stakeholders would have in this process.

I'm glad you mentioned this example -- I think it demonstrates 
exactly the concern I have.

You have overlooked an absolutely crucial distinction.  The
regulation of the pharmaceutical industry is done by the US
government.  The US government has, relatively speaking, infinite
resources available with which to enforce its regulation of the
pharmaceutical industry.  More important, perhaps: its regulations
have the force of law -- they are not merely clauses in a contract --
a different contract, potentially, with each company. 

Let's think for a moment about building an ICANN for the 
international pharmaceutical industry (PI)...the International 
Corporation for Pharmaceutical Industry Regulation (ICPIR)...

Let's think about replacing the current international regulatory 
structure.  We won't include regulation at the retail level -- that 
would plainly be impossible for a small, non-profit corporation to 
do.  Let's just think about the regulation of the industry itself.  
Let's think about the testing schedules for new drugs, as one small 
example -- enforced only through "reasonable contracts".  So ICPIR 
gets a complaint that a testing lab in Pretoria didn't follow the 
approved testing schedule.  Let's imagine how this tiny corporation, 
with a staff of 5 -- no, that's not fair -- let's say a staff of 
5000 -- are going to enforce this regulation.  They are going to fly 
to Pretoria and file suit.  Oh I forgot -- jurisdiction is not an 
issue -- by some miracle, all the drug companies in the world agree 
to subject themselves to lawsuits in the US for control of their 
business.

Also, remember that ICPIR is a *private* non-profit.  There is no 
law that says that the director of a *private* corporation can't 
receive gifts -- even enormous gifts.  There is no law that says 
that the Drug companies can't use whatever means they want to stack 
the elections for directors....

I find it hard to continue this exercise, to tell you the truth.  It
seems pretty plainly obvious that even the "minimal"(*) level of
regulation you mention for the drug industry would be impossible for
an entity like ICANN.  Even that "minimal" level of regulation
*requires* the force and resources of a government.  [I note that
some observers believe that ICANN *must* fail, and indeed, that
private regulation in general must fail, for exactly the reasons that
are brought out in this example.  I don't agree with that gloomy
assessment, but I do think that the problem is far more serious than
the glib free-marketeers believe.]

Personally, I think your "Pharmaceutical Industry" was a lousy
example.  Moreover, I think it is just absurd to speak of a "domain
name industry".  But many of the same concerns do apply. 
Opportunities for manipulation of ICANN abound; opportunities for
clever legal maneuvering abound; opportunity for political action
abound, and we don't have to look very far for proof: NSI really
isn't a very large company, compared to some of the drug giants, but
even so, there is absolutely no question that NSI is fighting a very
successful delaying action, employing some of these tactics. 

The point is, William, that it is absolutely insane public policy to 
create a regulatory agency with, relatively speaking, no power.  
There are two ways you can address this insanity -- you can give the 
regulating agency appropriate power; or you can manipulate the 
environment so that the power being regulated is not very great.

Now, some would argue that ICANN is not a regulatory agency.  But 
that horse is already out of the barn -- the agreements with the USG 
that require NSI to become a shared registry, and that put ICANN in 
the position of "NewCo", have created a regulatory regime.  It 
cannot be avoided.

================================================================
(*) some would not agree with you that Drug companies have mimimal 
regulation, of course.  But it was your example....

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