Re: FW: [wg-review] [DNDEF] short quizz 9,10

[Kent Crispin <kent@songbird.com>]

On Thu, Feb 08, 2001 at 06:38:27PM -0500, Joanna Lane wrote:
> Kent Crispin wrote:-
> 
> The changes I think need to be considered are:
> 
> 1) competition among arbitration service providers must be arranged so
> that they tend to produce uniform results.  "Forum shopping" should be
> restricted to issues of price and service, not anticipated results.
> [But if the trend noted by Mueller continues, and the numver

[Damn, hit the send before I intended.  The above sentence should read 
"But if the trend noted by Mueller continues, and the number of 
disputes continues to drop, this may be moot -- there may not be enough 
disputes to support multiple providers."]

> 
> Please explain. I don't understand how competition among arbitration
> services will guarantee uniform results.  (This is not a trick
> question). 

I didn't say "guarantee", I said "tend to produce".  If properly 
arranged, competition will tend to reduce variation in providers, just 
as competition in price tends to lower prices.

Right now there is a substantial difference in outcomes from one
provider to another -- relative to all other providers,
Disputes.org/eResolution strongly favor defendants.  This is a factor
that someone who was a defendant might consider very seriously, if
defendants got to choose provider.  But in the current scheme it is
plaintiffs that get to select providers, so plaintifs tend to favor
other providers, especially WIPO.  This is "forum shopping".  There 
have been several schemes proposed to alleviate the problem.  I favor 
one where the defendant and plaintif are required to rank all providers 
in order of preference, and the provider chosen would be the first one 
to appear on both lists -- if there are three providers, A, B, C, and 
plaintiff orders them A, B, C; and defendent orders them C, B, A, then 
provider B would be chosen.  If A or C consistently favored one side or 
another, then they would very seldom be chosen, because they would 
always be last on the other sides list.  A scheme like this gives 
business to the provider who stands out the least, in terms of having 
an obvious bias on any particular issue.

Note that this doesn't mean that the system would evolve to a position
where half the time the defendant won and the other half of the time the
plaintiff won -- it might reach a point where, across all providers, the
plaintiff won 90% of the time.  The plaintiff/defendant win-loss ratio
is not in any way a measure of the fairness of a system -- a perfect
justice system would catch every criminal, would never arrest an
innocent person, and the conviction rate would be 100% (since every 
defendant would in fact be guilty, because the system only got guilty 
parties... )

There are other schemes -- Louis Touton mentioned the simple scheme of 
having the pool of arbitrators be common among all providors, for 
example; Milton Mueller favors a "registrar decides" model, though I 
think that is plainly stupid, since it just moves the forum shopping to 
the registrar level.

> Also, value judgements on service include anticipating results don't they?.

If the system is set up so that providers have economic motivation to 
be objective as possible, then it would be very difficult to use 
results as a means to select provider -- at that point, issues like 
service and price take precedence.

> Or are you suggesting we specify service criteria?

No.

> 2) there needs to be a feedback mechanism that corrects for really bad
> decisions.
> 
> Excellent.

But very tricky.  Remember that one of the goals of the UDRP is to 
*not* have it be a substitute for the law, and if you build, for 
example, an elaborate appeals process, you are more and more making it 
into a legal system.

-- 
Kent Crispin                               "Be good, and you will be
kent@songbird.com                           lonesome." -- Mark Twain
--
This message was passed to you via the wg-review@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe wg-review" in the body of the message).
Archives at http://www.dnso.org/archives.html