Re: [arch-d] [IAB] The IAB Liaison Oversight program - feedback on closing

[Bernard Aboba <[e-mail]>]

Mark said:

"No, it means that those functions are disconnected from the program, and
have been performed by the IAB itself. We're currently examining how
everything is set up to make this more clear, and hopefully more effective.
In doing so, we're pulling in members of the ex-program who've expressed
willingness to lend their experience."

[BA] Heh. What is actually happening now sounds a lot like what was
happening when there was "a program" - in practice, the shepherds would
report to the IAB then too.

The "Program" structure was more necessary for the (somewhat separate) task
of "Liaison strategy",   Years ago, the relationship between the IETF and
some other SDOs was quite contentious, and non-IAB members were needed to
advise and to provide the history and how things got to the state they were
in and what should be done going forward.  However, most of the
then-important issues of "Liaison strategy" (like IANA strategy) resolved
themselves over time, and don't seem like as big an issue today.

On Tue, Feb 23, 2021 at 9:34 PM Mark Nottingham <[e-mail]> wrote:

> Hi Bernard,
>
> > On 24 Feb 2021, at 4:20 pm, Bernard Aboba <[e-mail]>
> wrote:
> >
> > I must say, I find this discussion quite confusing.
>
> I agree it's confusing, perhaps needlessly. This is at least in part
> because of how things are currently set up; see below.
>
> > The management of liaisons is a fundamental task of the IAB.  *How* the
> IAB organizes itself to carry out the task is up to the IAB.
> >
> > But the IAB does need to examine the functioning of liaison
> relationships on a regular basis, as well as monitoring the performance of
> liaison managers, so as to appoint or replace them as needed (or choose not
> to reappoint a replacement if there is no need).
>
> Indeed, and this has been done regularly since I joined the IAB, at least.
>
> > For that periodic review to get done, someone on the IAB has to step up
> and take a leadership role, and probably at least one other IAB member
> should take part, just to provide some backup/burden sharing.
> >
> > The individual(s) then report to the IAB on a regular basis.  You can
> call those individual(s) the "IAB leads" or a "Program", it doesn't much
> matter.  But you can't just say "the IAB will do it" - if you don't have
> specific individuals identified, and a regular rhythm for the work, then
> it's not likely to happen by accident.
>
> Yes. This has been done primarily by the Liaison Shepherds. During my
> tenure on the IAB (and to my knowledge for quite some time beforehand),
> we've also undertaken periodic whole-IAB reviews of the liaison
> relationships, to maintain familiarity and to assess the need for any
> action -- although this has fallen off somewhat during COVID. Additionally,
> the IAB as a whole often gets involved in specific liaison situations to
> assure that the appropriate parts of the IETF are brought into the
> discussion.
>
> There are some changes underway in how the Liaison Shepherds are
> allocated; the meeting minutes haven't been published yet, but essentially
> we're considering moving from a 1 shepherd-per-liaison model to a small
> pool of liaison shepherds responsible for the entire group.
>
> All of this is separate from the liaison program; it has not been involved
> in any of the above for the time I've been on the IAB, at a minimum.
>
> > And this regular work *does* need to be done - or else you end up with
> liaison managers who have long since lost interest, or worse - managers who
> take advantage of the lack of IAB oversight to overstep their roles.
> >
> > In that sense, liaison management is is quite a different IAB
> responsibility than say, the RFC Editor Program, where it was envisaged
> that the IAB would *not* get involved in the details.
> >
> > So when it is said "the program did not meet",  I am wondering whether
> this really means that the IAB hasn't been soliciting liaison reports on a
> regular basis, or paying attention to the performance of liaison managers.
> If so, that's not a statement about "the program" - it's a statement about
> the IAB - and something that should be brought to the attention of the
> nomcom.
>
> No, it means that those functions are disconnected from the program, and
> have been performed by the IAB itself. We're currently examining how
> everything is set up to make this more clear, and hopefully more effective.
> In doing so, we're pulling in members of the ex-program who've expressed
> willingness to lend their experience.
>
> Cheers,
>
>
> > On Tue, Feb 23, 2021 at 3:21 PM Wes Hardaker <[e-mail]>
> wrote:
> >
> > Here I sit.  Perched on the fence.
> >
> > I'm convinced that the existing program was either not needed or at
> > least not functional as is (not meeting is a clear indication of that).
> > Thus, I think a recharter likely wouldn't have been helpful because it
> > would have left an existing non-functional structure in place with a
> > minor incremental change (or at least, that's what I think the most
> > likely outcome would be).
> >
> > It would be best, IMHO, to decide what *is* needed and then come up with
> > a structure that works to support that.  I doubt it will look like what
> > it used to look like.
> >
> > Here's some example bullets of what might be needed for such a program:
> >
> > * Evaluation of the program as a whole
> > * Evaluation of liaisonships with each other organization
> > * Evaluation of changes for organizations with which we have or need
> >   relationships with
> > * Evaluation for how to best distribute incoming information from
> >   liaisonships to the rest of the IETF technical community.
> > * [your ideas here]
> >
> > Here's what should be out of scope (IMHO):
> >
> > * Evaluation of individual liasons -- that's in the IAB's charter
> > * Handling of the liaison reports/interactions themselves -- (again, IAB)
> >
> > Here's the things that might be needed to achieve the goals:
> >
> > * A person or group of people responsible to ensure the program (or
> >   whatever) is operational and active (chairs)
> > * Assured representation from the IETF (including, but limited to, the
> >   IAB, the IESG, the wider IETF community, ....)
> > * Assured representation from the external organizations that may wish
> >   to have a voice (the SDOs with which we interact may want to
> >   participate -- we may even facility dialog at a higher level)
> >
> > What are the risks?
> >
> > * this could quickly grow much bigger than need be (the above is already
> >   potentially too large)
> > * this could quickly grow too political
> > * this could quickly become bureaucratic
> > * And, like its predecessor, it could become forgotten and unused
> >
> > ...
> >
> > I think the liaisonships with other SDOs are critical to our success as
> > an SDO in the Internet ecosystem.  A lot has happened in the last 10
> > years that has proven that.  The existing liaisonships do work, and we
> > do hear from and communicate with and through them.  The question left:
> > is there more that should be done to improve the work, or better yet,
> > help the liaisons themselves be more effective?  If so, what?
> >
> > --
> > Wes Hardaker
> > USC/ISI
> >
> > _______________________________________________
> > Architecture-discuss mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/architecture-discuss
> > _______________________________________________
> > Architecture-discuss mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/architecture-discuss
>
> --
> Mark Nottingham   https://www.mnot.net/
>
>