--On Friday, March 29, 2019 11:39 -0400 Keith Moore
<moore@xxxxxxxxxxxxxxxxxxxx> wrote:
> On 3/29/19 3:28 AM, S Moonesamy wrote:
>
>>
>> According to the promotional material [1], using the Internet
>> is a good way to participate in an IETF meeting. What is
>> the IETF communicating to the persons who use that as a
>> means of participation in its "standards process"? It is
>> 2019; if one of those persons is of the opinion that there
>> is a justification for requesting a recall of an Area
>> Director, the person is unable to sign such a request.
>
> Your concern is valid and should be addressed somehow. But I
> can't help but have the impression that this is the least of
> the issues facing someone who wishes or needs to participate
> remotely.
>
> I see the recall process as a stopgap measure in case the less
> drastic and disruptive measures for ensuring fairness have
> failed. If recalls have to be considered more than once
> every several years, something else is likely to be wrong.
>
> So I guess I'm curious - what problems are we having that
> cause people to think we need to make recalls easier, and
> could those problems be addressed in less disruptive ways?
Keith,
Let me respond to your question with the understanding that my
answer may be very different from Subramanian's (or maybe not).
First, and most important, if one wants participation in the
IETF and for the results of IETF work to be seen as credible,
the processes have to be perceived of as fair and anything that
is offered as evidence of enabling people to effectively has to
have the appearance of being workable. From that standpoint, if
we need a recall process at all, it must be available to all
participants in the IETF. If the definition of "participant"
for initiating recalls is different from the definition for,
e.g., making a comment on a mailing list or submitting an I-D,
we'd better have a good explanation for the difference. If we
don't, the odds of someone concluding that the IETF is biased
toward people with high levels of travel and related support and
with sufficient time on their hands to be away from the
workplace for three or more weeks a year to attend meetings are
very high. That perception of unfairness may discourage
participation as well as the perceived value of IETF consensus.
That is the case whether we actually use the recall procedure
every year, every several years, or, in practice, not at all.
Comparing recall initiation to, e.g., serving on a nomcom or
recall committee, if is possible to see unfairness in those
elements of our leadership selection process as well. For
example, in an ideal world, we would almost certainly want to
have people who participate only remotely included in the
volunteer pool for the nomcom. We don't do that. However, we
can explain, in terms of the way the nomcom works in practice,
why f2f participation is important. Do I, as a mostly-remote
participant like it? Nope. The the argument is plausible
enough that I don't feel it is an unfair procedure that both
unreasonable and easily corrected.
Beyond that, there are several potential sorts of problems for
which we have no "less drastic and disruptive measures for
ensuring fairness". Let me give two examples.
We've seen the first in the form of a situation in which someone
stops performing in their roles and either will not resign or
will not or cannot respond to requests that they resign. Recall
is then the only mechanism we have for getting them out of the
position and getting someone who have the time and resources to
do the job into it. Maybe we should have some other procedure,
e.g., one in which certain types of repeated non-response create
an implicit resignation but, at the moment (and when we nearly
had to use it) recall is the only mechanism we have.
For the second, borrowing a bit from an off-list discussion, and
with the understanding that this example is completely
hypothetical, suppose someone in an AD position started
expressing very strong opinions about choices of technical
outcomes, perhaps even as the relevant AD (for which that person
was responsible) was just starting its work and being repeated
when alternate approaches were advocated by others. Also
suppose that type of presentation appeared to be turning into a
regular patter of behavior by that individual. Now, perhaps
people who found themselves on the other side could wait until
the work in question made it to the IESG and into IETF LC, see
if the decisions made there included that AD's taking a
position, and then appeal a decision that aligned with the view
expressed earlier and keep repeating that for every other
decision in which there had been a strong AD intervention early
one. However, especially if there is a pattern and other than
trying to have a conversation with the individual and/or
encouraging the IETF Chair or other members of the IESG to do
so, I suggest that a recall would be the right answer, if only
because waiting for the WG's results to reach the IESG and IETF
Last Call is the wrong way --and much too late-- to handle the
problem without damage to the IETF's technical work and
consensus processes.
In either of those examples, "something else" is clearly wrong.
In the former, one might suspect illness or personal issues well
beyond the ability of the IETF to intervene in a positive way
other than letting the recall process run its course. In the
latter, a situation like that would probably indicate a failure
on the Nomcom's part in making the appointment in the first
place, but, assuming informal conversations failed, the recall
process would likely be our only effective mechanism for getting
the relevant person's attention or, if that was not successful,
in getting the behavior stopped.
best,
john
