Re: [wg-c] "It Won't Work"

[Jonathan Weinberg <weinberg@mail.msen.com>]

At 11:16 AM 4/11/00 -0700, Dave Crocker wrote:
>This note raises a question about Working Group C process.  As such, I am 
>requesting a formal response.
>
>For development efforts, there are two kinds of disagreements:  One is "I 
>don't like the idea" or "I prefer some other idea".  That is an argument 
>along the lines of engineering preference.  Hence, consensus is a 
>reasonable way to break a deadlock, although it does raise a question about 
>relative qualifications for choosing.
>
>The other type of objection is "It won't work".  In other words, it is a 
>claim that people's preferences are irrelevant, since the proposed 
>mechanism simply does not function as desired.  To deal with this concern, 
>there needs to be an analysis that shows that the mechanism DOES work, or 
>else the mechanism/proposal needs to be fixed.
>
>A number of us have raised concerns about the list of "criteria", pointing 
>out that there is no objective way to apply them.  As such, they will 
>become yet-another source for subjective debate.
>
>In other words, the concern about the criteria is that They Won't Work.
>
>What is significant is that there has been no substantive response to this 
>concern.
>
>Hence my raising a process challenge.
>
>In other words, the current consensus call is irrelevant until basic 
>concerns over the ability to apply these -- or any other -- "criteria" are 
>resolved.
>
>Jonathan, please deal with these concerns explicitly, since they concern 
>the ability to USE the "criteria".  If there is no clear and reasonable 
>basis for applying the "criteria" then development of them is just an 
>academic exercise.


	The gist of the response as I understand it (and in my own words) is this:
Any set of substantive standards, offered up to guide a decision-maker, is
going to fall somewhere on a spectrum between 1.00 (totally constraining:
the decision-maker applies the criteria in an entirely mechanical way, and
that process generates unambiguous instructions about what to do), and 0.00
(totally irrelevant: the decision-maker, after attempting to apply the
criteria, has absolutely no more guidance about what to do than he had
before he started).  I understand the proponents of the S/K principles to
argue that the S/K standards fall in between: that they will offer some
useful policy guidance to ICANN even though they're not totally
constraining.  (Philip has recently suggested that later in the
decision-making process, an ICANN body may wish to take the general policy
guidance embodied in S/K and embody it in a more specific implementation
document.)  Notwithstanding that the S/K principles will not themselves
generate a mechanical, objective (>.75 on the scale?) process, proponents
argue, the policy guidance they provide is still valuable.

	Each member of the WG, natch, has to decide whether that response
satisfies them.  In order to conclude that the principles will "work," as
Dave puts it, one has to conclude [1] that the principles do provide *some*
(>0.00) policy guidance, rather than being totally irrelevant to actual
decisionmaking; [2] that the policy guidance they provide is desirable
(that is, that they move in the substantively correct direction); and [3]
that the advantages of the policy guidance they provide aren't outweighed
by any disadvantages they may have (e.g., providing another hook for
lawyerly types to try to tie the process up in knots).  But all this is
stuff we've been talking about, and I assume that people will vote "yes" or
"no" depending on their understanding of those issues.

Jon