Re: [IETF] Re: Appeal Response to Abdussalam Baryun regarding draft-ietf-manet-nhdp-sec-threats

[Pete Resnick <[e-mail]>]

On 7/3/13 1:10 PM, John C Klensin wrote:
> --On Wednesday, July 03, 2013 13:02 -0400 Warren Kumari
> <[e-mail]>  wrote:
>
>    
>> Thank you -- another worthwhile thing to do is look at who all has appealed and ask yourself "Do I really want to be part of this club?"
>>
>> Other than a*very*  small minority of well known and well respected folk thehttp://www.ietf.org/iesg/appeal.html  page is basically a handy kook reference.
>>      

I think this is a bit overstated. There are 14 unique names of 
appellants (2 of which are groups of appellants). As I stated, 3 of 
those appellants account for 19 appeals, all denied. Perhaps you don't 
want to be part of the club with those 3 who make up 60% of the 
appealing, but if you simply remove those, you get:

13 appeals for 11 appellants (2 of them appealed twice, with years in 
between appeals)
1 appeal withdrawn before the IESG decided
6 appeals accepted
6 appeals denied.

So the small minority are actually the repeat appealers. Of the rest, 
over half I would instantly recognize as well-known and long-time 
participants, and (without naming names) half of *those* folks were 
denied and half were accepted.

So appeals that get to the level of the IESG from the group of 11 are 
accepted half of the time. That means that these folks are bringing 
issues to the IESG that, after having gone through the WG, the chairs, 
and the cognizant AD, half the time are still accepted by the IESG. That 
is, there's a 50/50 shot they've found a serious problem that the IESG 
agrees the rest of us in the IETF have missed.

I'd be part of that club.

> I am honored to be a member of that club.   Remembering that
> appeals, as others have pointed out, a mechanism for requesting
> a second look at some issue, they are an important, perhaps
> vital, part of our process.  We probably don't have enough of
> them.  Effectively telling people to not appeal because they
> will be identified as "kooks" hurts the process model by
> suppressing what might be legitimate concerns.
>    

Agreed. In any dispute process, there will be some folks who are 
outliers that make up an awful lot of the total load. But that shouldn't 
take away from those who are using it for its designed purpose.

> In addition, it is important to note that the page does _not_
> list every appeal since 2002.  If one reads Section 6.5 of RFC
> 2026, it describes a multi-step process for appears in each of a
> collection of categories.  The web page lists only those that
> were escalated to full IESG review.

Interestingly, 2026 6.5 only refers to things that get to the IESG, IAB, 
or ISOC BoT as "appeals". The rest of the "discussions" are simply part 
of "dispute" or "disagreement" resolution.

But John's central point still stands: Most of the dispute resolution 
takes place before it ever gets to the IESG, IAB, or ISOC BoT as a 
formal appeal.

> p.s. to any IESG members who are reading this: community
> understanding of the process might be enhanced by putting a note
> on the appeals page that is explicit about what that list
> represents, i.e., only appeals that reached full IESG review and
> not all appeals.
>    

Good idea.

pr

-- 
Pete Resnick<http://www.qualcomm.com/~presnick/>
Qualcomm Technologies, Inc. - [phone]