Minutes IETF98: iasa20
||Minutes IETF98: iasa20
IASA 2.0 IETF 98 Vevey 1 & 2
# Joe Hall - Report from Virtual Workshops
Joe described how he and Jean tried to organize issues and rank them
in terms of importance.
Joe shared an overview of the organizational structure (too many boxes
here to summarize).
A number of structural issues were described, particularly the
relationship between IETF and ISOC. Joe identified several issues
with the delineation between the organizations.
Issues with funding sources for IETF activities were outlined. Joe
provided pointers to drafts by Jari and Leslie on these subjects.
The staffing issue was raised.
Finally, the internal IAOC organizational structure was mentioned.
# Olaf Kolkman - ISOC Contributions
ISOC gives ~ $2.3M (2017) and covers IETF budget shortfalls on
occasion. About the same amount comes from the other IETF revenue
sources. ISOC charges some of its activities back to the IETF.
Olaf described the ISOC community diversification and outreach
programs, including the IETF journal, ARNP, mentoring. Regional
diversification. Miscellaneous other things.
# Jari Arkko - IASA Challenges (and Opportunities)
Jari made some observation on finances, its health and trends
affecting it. He noted that this indicates a need for change.
Jari observed that the structure had some problems and he identified
some specific problems.
# Alissa Cooper - Summary Statement
Alissa expanded on the complexities that were not captured in the
schematic we were shown.
Alissa suggested that we keep funding and administration separate.
Alissa contends that the structure doesn’t allow for the sort of
accountability we might want.
Alissa asked that people try to help formulate ways of approaching
solutions to these problems.
# Mic Line
Lou Berger - Where does the Trust fit in?
Alissa - The trust is obviously implicated by this. There simply
wasn’t enough room on the slide to include a box for the trust. The
report from the workshop includes it.
Lou - The trust should be part of the discussion.
Alissa - Definitely.
Andrew Sullivan - Do we make tiny changes, or a big change. The
structure is too weak. It needs very, very significant changes. It
overloads a number of positions in a way that it shouldn’t. Someone
involved in the original design said that “It has to be the IAB chair
because they have all the state”. My response is that this is a
weakness and we should not route all the one person - we would reject
a technical architecture that had a magic box at the centre that did
Ted Hardie - Focus on the single individual is a risk. The magic box
issue is core to this. Is this the right set of connections. I say
no. All of our data is coloured as a consequence of the structure of
the organization. We need more than a refactoring, but need
restructuring. This is likely a lot of work.
Harald Alvestrand - I like hearing my own voice. The trust was thrust
upon us from certain quarters, and we could have added another label
to the IAOC. We wanted for a few things to not happen when we set this
up. “If you have to do work, then you are doing it wrong.” was a
principle. We put one person in to handle contracts. Then that
person needed to take orders from someone, other than the community as
a whole. So the IAOC was created for that purpose. We did not want
the IAOC to become a power centre, so we put the chairs in so that
there was no question about it being a power centre, it was another
chore for those chairs.
Jonne Soininen - We were asked if we should consider small tweaks or a
bigger overhaul was asked. Maybe we should talk about the target and
let that drive the decision. We need to ask what organization we might
want. It’s a very different scenario to what we had in that time. We
need to ask how the IETF has changed over those years. Where we used
to ask for volunteers, we find more professionals being involved.
Jon - What do you think the target should be?
Jonne - Some of the things were done in the design of the time,
looking at that does that still fit. I don’t know what the target is.
The situation warrants review. It’s 12 years and the target we had 12
years ago doesn’t exist anymore (“do we still have the fate in our
Jon - There is some question about whether we do have the fate in our hands.
Cullen Jennings - Speaking for Ward. We sponsor both the ISOC and the
IETF. We would like those to be very separable. That’s how we manage
things internally. We would like to sponsor the IETF. We would prefer
entirely separate organizations that work together.
Leslie Daigle - The reason that the IAB chair is on the IAOC is that
they are plugged into all the IAB activities. The IAB chair and IETF
chairs are two peas in a pod. I don’t think that solving the
overloading of those individuals is a responsibility of the body they
represent. I was the first IAB chair on the IAOC, but then had a long
gap before I later joined the IAOC. In that time, the world changed a
lot. We did not have the RFC editor contract, we were only starting
with the secretariat contract. We did a good job in 2005, but that
doesn’t need to take anything away from looking at this carefully. I
think we need significant changes. We didn’t have experience in 2005,
we have a lot more now in those more mundane things.
Lucy Lynch - Echoing Harald and Leslie. We need to get a level above
this and look at the reasons for the design. What do we still want.
I still want us to be mostly volunteer. We need an independent
relationship with our funding sources. How many organizations
reorganize every 6 months.
Jason Livinggood - Echo Cullen. Far easier to have separation. Far
easier to explain to finance and accounting. I would echo Lucy, 12
years is a long time for any organization. If this were an aeroplane,
we’d be at the outer limits of the design envelope. We need a clean
slate design. The time is right to do that now.
Bob Hinden - This chart reminds me of the analogy of paving the
cowpaths. The structure we have has evolved. The original design was
successful in that way. I’m sure we could do it better, but we need
to think hard about what sort of organization we want to be. Going to
paid staff isn’t a good idea, because those staff will run the
organization and the community has less of a say. See how the W3C
operates, which is the other extreme, and they are having
organizational issues. We wanted to be self-governing, and we got
that, but we struggled to get enough volunteers to run this process.
Eric Rescorla - Where is the P-CSCF? I was there when this was
designed. It as a reasonable design at that time given those
constraints. The principles haven’t changed, but the environment has.
It’s clearly time for some sort of refactor. There isn’t much danger
of the paid staff running things. The W3C have paid staff as part of
the technical mix and the secretariat don’t do that now. It would be
much better (as Cullen said) as a donor, to have separate pots for
specific things, rather than donating into some global pool that is
Kathy Brown - I sign everything and am legally responsible. The
workshops were excellent. ISOC has deep commitment to the IETF. We
feel that we are partners. There are deep-seated relationships. This
structure is odd. You have an activity inside a company that exists
inside of a non-profit. The IETF rightly wants to be independent and
also sits inside an organization that makes its own legal decisions,
which complicates things. ISOC are fully committed to be partners as
you explore this issue, including the big funding issue.
Alissa Cooper - I want to come back to some issues. Volunteer-lead
activity. Agree with ekr, we can rely on this remaining
administrative only. It seems quite clear that the bulk of the
community isn’t interested in taking on the bulk of the administrative
tasks. If we are looking to attract the high quality engineers, those
people won’t be suited to admin tasks, nor will be they be inclined to
do that. The participant base isn’t particularly well-suited to that
task. Independence from funding sources. Are we independent from
Lucy - Distinguish between the volunteer org and orgs run by
volunteers. We need to be able to have volunteers to influence the
process. Network setup is volunteers and paid staff, but we have what
we have because volunteers are active in the process. We want
self-determination and we need smart hands to help accomplish those
tasks. On funding, I would have to think about that. When we did
this in 2005, we wanted to ensure that we never had a partnership
arrangement rather than an ownership arrangement. Now we have a
partnership that needs renegotiation.
Leslie - Eric captured my thoughts on the volunteer side of things. I
appreciate how Lucy is characterizing things. We need to rearrange as
necessary. We can’t let that preclude us organizing as we see fit. I
would like to know who is directing our communications plan. It’s not
an IAOC thing. The IETF chair has to do that now. The funding
challenge is that we understand how we represent ourselves so that
others who might engage can understand it so that they can contribute
time effort and money.
Jari - No one is suggesting that the community would lose control, but
more that we have imperfect tools for that. I think that we will
still have a board, and they will still be Nomcom appointed. What is
the goal? It could be the same goal as what we had way back: gain
control of our administrative destiny. Let’s do the resign so that we
keep the control. We’re already paying people to do this work, but we
Dave Crocker - There are lots of issues, which could be overwhelming
and we could do badly if we don’t organize the ideas. We should look
for things not to do. e.g., People over in ISOC that we don’t manage.
In some cases, it might be valuable to be involved, but others not.
We don’t do performance evaluations for AMS staff, we provide input,
but it’s a contractual arrangement. We hire ISOC for several tasks in
the same way. ISOC is family and family arrangements are interesting.
It always produces unexpected effects.
Jon - We should talk about the target as a way to overcome the mess of problems.
Paul Hoffmannnnnnnnn - Communications struck a chord with me. If we
want the outside world to engage, we need to invest in telling the
world about what we do. ICANN doesn’t understand us very well, for
example. Think about how we communicate this to the community, who
might not be engaged in this process, but still care that the IETF
continue to exist.
Ted - My target here is an effort to change these interfaces so that
the people who do the tasks, are doing the tasks that they are best
suited to do. We wasted two years of Andrew’s talent as a former IAB
chair. We wasted his time on something for which he was not selected
for. This hasn’t been good for our organization, or the individuals.
Refactoring the organization so that we can make selections that fit
Pete Resnick - This started in 2003/2004 as a problem of family. We
had a falling out with the part of the org that did admin. We don’t
want that to happen again. No one thinks of the IETF in the way that
Kathy described. We need to restructure this so that we are clearly a
family but so that we don’t have any part subservient is important. I
have no problem with professional staff doing admin, as long as we
have people who are able to oversee this.
Olaf - Everyone at the mic was around in 2005. Where are the new people?
Glenn Dean - As a global host separating the names on the cheques
would definitely make this easier. There is a shift in the industry
that is leading to declining attendance. The pool of volunteers is
shrinking. We can no longer make up the gap. The number of people
here is dwindling. We need to execute the administrative side so that
onboarding is much easier so that we attract and keep new people.
Joe Hildebrand - I don’t count as new any more. We have a lack of
bench strength for leadership positions. We need to accept new work
to do that. We need to own our part in how we treat people who come
to use with new work.
Alissa Cooper - I wasn’t here in 2005. But I appreciate the
perspectives of those people who were. Some other principles
regarding the target. Most people who participate in the IETF just
want this to work, they don’t care how it is done. And that is fine.
We need to ensure we get as much input as we can, but can’t expect
those people to provide much help. We need to ensure that what we
produce is governed by the community. We need to ensure that we can
continue to raise money and attract new participants to the IETF.
These are all areas in which we can improve upon.
Richard Barnes - Also post-2005. I find this baffling, but the family
metaphor works. Lots of arrangements are informal in a family. I
write a contract when I loan money to my family. Making good fences
makes for good neighbours. We need to know where those fences - the
needs - are. That is the first step here. I don’t have a good answer
Alia Atlas - I was here in 2005, but wasn’t involved in IASA. We have
lots of ways to grow technical management (chairs, ADs). We do
apprenticeship and mentoring there. We have nothing for these
administrative tasks. A lot of our community includes senior
engineers with management experience. If we don’t nurture and mentor,
we won’t get people. I also share the need for outreach and growth.
We’re losing institutional knowledge as people retire.
Ray Pelletier - I am the IAD. HELP! We need to look at the IETF
trust. I can’t watch everything, which is why I suggested we split
IAOC chair from the trust chair. The IAOC always has trouble
selecting a chair with so many ex-officio members that we have very
few valid candidates. Mission critical support is what we do, like
Meetecho and tools development. I think we have done a pretty good
job. Sometimes we step on toes, but that is why we have oversight
from the community. Every time I ask the IETF: what do you need? I’m
not getting an answer. … IETF is an activity of ISOC. I had a
similar arrangement, where the activity had a divorce from its parent.
That was successful, they got more attendance, more money and were
more successful. … I don’t want to die on the job, I don’t want to
die from the job. So help.
Dave - Asks for hands >10y: lots. 3y were mostly ISOC staff, and Joe
Hall. We need to avoid delegating our culture so that we maintain our
culture. Sometimes we don’t delegate where we should, we’re not
always good at that.
Eric - I’m hearing a lot of the same things from a lot of people. I’m
looking to close out .
Gonzalo Camarillo - The target should be separation of concerns.
ISOC/IETF relationship is confusing. The ISOC board is working on
this. We have trouble explaining this to the non-IETF members on the
ISOC board. ISOC is looking at reorganizing to improve how they can
better support the IETF. The split is hard to explain. My budget at
Ericsson goes to ISOC, but ideally some other division would send
money to ISOC and my budget would go to the IETF. Let’s get to next
Randy Bush - Newcomer badge. As the Internet becomes successful it
becomes vendor backed. It all comes down to economics. Sending
someone to attend on a finance committee is hard to justify. We
pretend that this ISOC/IETF relationship is a partnership, but without
the IETF, ISOC doesn’t last 3 years. We need to take more
responsibility at the IETF for the things that ISOC is doing. We
don’t even know if these things happen until Olaf tells us.
Alissa - Thanks for being frank everyone and helping. I heard a lot
of the same things over and over. People seem to be roughly aligned
around the framing questions. It sounds like there is openness to
changes. ¾ of global hosts mentioned that. We should continue to
talk about Jonne’s question about targets, on the list. We can at the
same time think about the shape a solution might take. I think we
should continue to run this in a workshop style and keep momentum up
on the list. In terms of driving, this is on me, but I will be
reaching out to people to ask for help.