ietf
[Top] [All Lists]

Re: Out-of-area ADs [Re: IETF areas re-organisation steps]

2014-12-28 17:13:24
On Sun, Dec 28, 2014 at 11:15:06AM -0500, Pete Resnick wrote:
On 12/26/14 1:51 PM, Brian E Carpenter wrote:

Do note the explanation for this in our original message, as I think
you have an embedded assumption that's at least non-obvious:

There are numerous instances where the constituency of a WG exists
in a particular IETF area, but the most appropriate AD for that work happens
to be in a different area, or where the ADs in the area are simply overloaded
and an AD outside of the area is perfectly capable of managing the work.

To expand a bit: We do think having WGs in particular Areas is
useful, and we think generally we're getting our assignment of WGs
to particular Areas correct: At the scale of an Area, it is
generally quite obvious and natural that the kind of work we want
done in a WG falls to a particular IETF constituency, and those
folks are normally in a particular Area.

However, we have not found that ADs are so specialized that there is
a "correct" AD for every WG, or that the AD whose main area (or
Area) of expertise is always the best person to manage any
particular WG. [...]

We think the current way we've been doing assignments are a bit too
rigid. It *should* be normal for us to assign specific WGs to the
best AD for that WG, even if the best AD for the Area happens to be
a different AD. That flexibility should let us redistribute the load
as needed, and hopefully make it easier for the NomCom to fill
slots.

This.

For me then this comes down to: once-in-a-while "restructurings" that
consist of one-but-more-likely-many WG/AD reassignments, versus more
frequent one-or-two WG/AD reassignments.

For me "flexibility" as to WG/AD assigments, among other things, means
that each such assignment can't be allowed to require IETF consensus
with 100+-post threads on ietf(_at_)ietf(_dot_)org.  To me WG/AD assigments 
should
be a matter strictly for the IESG and WG chairs; WG participant and IETF
input should be welcomed, but their consensus should not be formally
required.

The alternatives are once-a-year restructurings yielding long
ietf(_at_)ietf(_dot_)org threads (OK), less frequent restructurings (not OK), or
death by a thousand 100-post threads.

I think[*] I much prefer frequent one-off WG/AD reassignments, _without_
the overhead of IETF consensus in each case.  Someone is bound to see a
conspiracy in that view, which is probably why this isn't on the table :(

Nico

[*] Eh, I can't be too certain about it; I've no idea how that might
    work out IRL.  Maybe we'll hate the result, but we can always go
    back to the alternatives later.