Re: [weirds] fyi: WHOIS Policy Review Team Final Report

[Andrew Sullivan <[e-mail]>]

On Thu, May 17, 2012 at 01:33:57AM -0000, John Levine wrote:
> This is a surprisingly good report.

It's interesting you say that.  I had exactly the opposite reaction.
There are problems both large and small.  The executive summary has a
bad definition of what a domain name is.  The report makes a number of
preposterous recommendations, like that ICANN should, for some reason,
start chasing all referrals in whois (because, apparently, people
don't know how to use whois and this will magically enable them to do
so), and attempts to extent ICANN's regulatory reach into areas where
it has no business (and where it will fail anyway).  It gets its
history wrong, and just ignores IRIS.  It makes the distinction among
the service, protocol, and data but does not attend to that
distinction throughout.  It points out, but offers no suggestions for
resolving, the basic inconsistency in what different communities want
from the registration data service.  Finally, it simply refuses to
engage with the question of whether the very limitations of the
protocol are a fundamental part of the problem.

The latter is the most serious issue, in my opinion, because it leads
them to make recommendations that are just as unrealistic as the last
five times ICANN has blathered on about whois.  Fixing the protocol
limitations is simply a necessary condition for doing anything about
all the rest of it.  I sent them a public comment pointing this out
after they posted their draft report (I also sent them private mail
pointing out the number of technical errors in the report, most of
which they appear to have left alone.  One sometimes gets the feeling
that ICANN committees just don't care about technical precision, and
this report doesn't help dispel that feeling).

I think the report is a shame.  It has taken several years and not
insignificant money to say a bunch of commonplaces, yet the report
doesn't really help do anything about the two most serious problems
with registration data: the protocols we have are poorly adapted to
serving the needs we have, and the set of needs we have is in any case
an internally inconsistent set.  The first is a technical issue, and
we here are in a position to do something about it if only we
understand what problems we need to solve.  The second is a basic
problem of public policy, in which different actors want vastly
different things from the same service.  One might have hoped that the
report would have provided a framework for figuring out how to make
those compromises, but it doesn't.

> I'd encourage people to read it,
> at least the first section which summarizes the recommendations, and
> send a comment to ICANN.

On this we agree.

Best,

A


-- 
Andrew Sullivan
[e-mail]