Re: [wg-d] WG Principles

[Karl Auerbach <karl@CaveBear.com>]

> In any case, the "legislative history" clearly indicates that IETF 
> processes are an underlying model for the development of policy in 
> the DNSO

Being the only person on this mailing list who has been a chairman of an
IETF working group (not to mention that I am also married to one and work
daily with another):

The IETF working group model is entirely inapprpriate for the discussion
of soft policy issues.

I know.  I was co-chairman of the IETF Poised (policy and procedures)
working group and I watched it struggle, and it continues to struggle, to
reconcile substantial differences in opinion.

Soft policy, as opposed to technical development, requires a more
formalized discussion structure, with many external checkpoints along the
way to validate that the working group has not lost its way and become
isolated.

Soft policy, as opposed to technical development, requires that the
working group product be considered not an end awaiting rubber stamp
ratification, but merely a possible starting point for acceptance by
broader and broader groups of interested parties.

In other words, DNSO working groups are a place where policy *proposals*
can originate, but the results should be considered merely proposals that
can be rejected or completely rewritten.

And unlike the IETF, in which working group proposals bear a presumption
of validity and thus tend to march along unless problems are shown, DNSO
working group output should be presumed to be simply tenative and ought to
have to be accepted by increasingly larger circles of interested parties.


By-the-way, I disagree that Bret's principle's represent the IETF model at
all.

Bret's model is one in which the DNSO delegates work downward to working
groups. In the IETF working groups tend to bubble up from the bottom,
IETF working groups are often born from "BOFs".

IETF working groups are the primary vehicle of IETF work, they are not
subordinate to the IESG or the IAB.  Yet in the DNSO, all working group
output should, and must be, subordinate to the opinion of the General
Assembly.

Matters in IETF working groups are susceptable of objective evaluation and
are never settled absent implementation experience.  In DNSO working
groups, the criteria are subjective and there is little or no opportunity
for "implementation experience."

The IETF working group model is fine for relatively focused technical
matters among people with relatively well aligned mindsets and goals.
But it is not a good model for the development of domain name policy.


		--karl--