Skip to main content

Minutes for IESG at IETF-87

Meeting Minutes Internet Engineering Steering Group (iesg) IETF
Date and time 2013-07-31 15:40
Title Minutes for IESG at IETF-87
State Active
Other versions plain text
Last updated 2013-08-02

IETF 87 Plenary Notes
Note taker: Andrew McGregor

1740-2010  IETF Operations and Administration Plenary

   1. Welcome
   2. Introducing the sponsors
   3. Reporting
      - IETF Chair
      - IAOC Chair and IAD
      - IETF Trust Chair
      - 2012 NomCom Chair
      - 2013 NomCom Chair
   4. Postel Award

Elizabeth Feinler

   5. Recognition

Hugh Daniel and Evi Nemeth

   6. Diversity Team

x: I support the effort, proposed a specific nomcom question on specific
intercultural skills.  IESG function leads to self selection, please clarify?

Suresh: Contention is the IESG is not diverse, and gatekeeps future work,
therefore selects against a certain amount of diversity (because some ideas
seem irrelevant)

x: Not sure if I agree, but thank you for the clarification

Suresh: This is community feedback, required more than one mention.

y: Until recently involved with SIGCOMM, we have had very similar discussions
in CS and ACM community, we really tried hard to address diversity issues. 
There was a published blog in communications of the ACM, plus some concrete
proposals.  We don't need to reinvent the wheel here, unfortunately it's a
common problem.  Secondly, the time involvement is a considerable barrier. 
Many people do not have the time to bring a document to completion, so making
that time more predictable; for academics, probably beyond the period in which
they are relevant.

Klensin, via jabber: Why could we not have these slides uploaded in advance,
that would help?

Jari: My bad

Keith Moore: I mostly agreed, but someone could not deal with the IETF
culture... however, what does that mean?  Some aspects are significant to the
success of the organisation, for example the 'good of the internet, not the
interests of corporations'.  Some people's interest is not good for the

Kathleen: As strong as I am, as a woman in this field, I sometimes took a day
or more to respond, and I have heard from women who felt less excited about the
organisation because of the (email) culture.  We certainly need good technical
discussion.  There are research culture differences world-wide yet they get
work done.  Technical discussion is essential; however, sometimes when people
hear a very forceful statement, it is merely for precision and not personal.

Jari: Keith, one observation is that this is not black and white.  We can
somehow improve the IETF, without abandoning the basic principles.

z (asian accent): Diversity was a controversial topic two years ago, and I am
happy to see that things have somewhat changed, in that the leadership is more
diverse and there is a diversity project in action.  The IETF is about
connecting networks, and bringing in people from more parts of the world will
help achieve the IETFs goals.

a: Thank you for this presentation, and had you asked me in Orlando if we would
see significant progress in three months, I would have been sceptical.  But we
have, congratulations.  Diversity gives better results, and as engineers we
should appreciate.  It isn't necessarily what you say, but how you say it, and
that's one of the things we should work on.

b: Is there a document that explains how mailing lists work, that explains the
social framework and that people don't necessarily mean things as badly as they

Suresh: The Tao is probably what you want to read.

Kathleen: There is a video in production that will help explain these things.

Harald: Sorry about making your jobs harder.  We are an open community, which
is one of our core strengths.  Which also means, some members do not share the
community goals including diversity.  Which means that our leadership, wg
chairs, IESG and IAB, need to take action when needed to stem actions that
would lead to a reduction in diversity.  I'm a bit worried about the assertion
that if we were a bit nicer, it would go away; there will be occasions where we
need to be a bit more brutal with some of our members.

Jari: In many cases we have chosen to not get in to these arguments, but I
believe we do need to take them up.

Paul: Suresh, I'm the editor of the Tao, and we don't discuss much about the
culture of the mailing lists.  We tried to put that in 5 or 6 years ago, and
tried to explain it.  We were asked to take it out, since it seemed unfriendly
and would scare people off.  I would love to put it back.  Sometimes our
mailing lists are a bit obnoxious, I'm one of the culprits.  I would like to
see it addressed.

Kathleen: If it is too much to put in directly, recommend tips such as being
persistent.  These things are hard to wordsmith.

Paul: If part of what the design team wants is to introduce more to newcomers,
the Tao is still the gateway, and could have some edits.

Kleinsin (via jabber): <missed>

Tobias: My first wg meeting was horrible, I felt I received all this bashing
(rightfully, to some degree).  I would like to reinforce, it's about how you
say things, not what you say.  I've travelled globally, and you can raise the
same issues in different ways.  I would like to push back on training
newcomers; we also need to train the long term participants, it is for us to
adapt to how the rest of the world is as well.

Keith: I somewhat object to that last statement.  Certainly, how things are
worded does make a difference; everyone colours statements in their own way,
based on their experience.  I would argue that we need to be careful about
expecting people to be too careful.  I have seen what happens when you put the
responsibility on the speaker about how his words are received, and the result
is not pretty.  Be conservative in what you send, and liberal in what you

Suresh: Many people do not perceive that they are being aggressive, and can
easily change behaviour.  If you can be nice, why be a jerk?

Keith: ... <missed>

   7. IAOC Open Mic

Jon Lee: I realise VAT is your least favourite word... how much do you expect
to get back?  It looks like the income and expenses are about equal, and should

Bob: Ray has looked at it, about half of what we have to pay will come back

Ray: About 132k to pay, recovery 60k to 70k

Adrian Farrell: I'd like to volunteer to help post things before the start of
the plenary

Bob: We try to make the financial data as current as possible, the rest does
not change as fast

Eric: Cookies joke... off-the-wall thankyou.

Bob: All credit to the Secretariat


Tim: I believe the IAOC remains sensitive to increasing fees.  VAT was an issue
because it was too late to increase the fees.  I hope you will continue to be
sensitive to increasing the cost of attending, sum of fee and hotel cost.  I
would hope that additional taxes would become a factor in choosing venues.

Randy: We are already locked in to places that will have VAT.  I fear that this
is going to be more and more places, not just those that we have picked. It is
an unanswered question whether we eat that, or whether we say the fee is what
it is, and the tax is extra.

Bob: The EU passed the general rules, countries implement them, and therefore
this may vary by country.  There will be ongoing activity to understand this,
and if the rules treat us differently, we have to take that in to account.

Ray: The IAOC considers the fee every year, it has been $650 for three years,
and we don't frequently increase it.

Scott:  The ISOC board specifically funded keeping the fee stable some years

x: VAT is not new, in previous years may have flown under the radar.  This may
not continue, as the regulators are getting stricter.  At the end of the day,
VAT is meant to be neutral to organisations.

Bob: We believe the rules changed in 2011, and there are some meetings in an
ambiguous state, which is not clear at this point. We may need official EU

y: IEEE is having the exact same problem, and it's a pan-European problem.  We
will find that wherever we go in the EU, we'll have similar problems.  I do
want to recommend that it would be helpful to consult with the IEEE

Bob: We have started that, and will do more

Pat Thaler: Vice-chair of IEEE 802.  We have had IEEE legal and accounting look
at it, and the opinion was that conferences had to pay, and standards meetings
did not.  We looked carefully.  I don't think anyone in either organisation
wants to get in trouble.  IEEE legal has been asked to look at the regulation
change as well.

Steve Hanna: Not a VAT expert, but feedback on whether IETF should eat the cost
or pass it on, and I believe you should pass it on, because at least some large
portion of us will be able to have our employer get a VAT refund.  It does not
make sense for the IETF to eat it when it can be retrieved individually.

Bob: We'll consider that in the budget

z: As a self-employed person, we do not need to bias even further against
individual attendees.  I can't remember a better meeting site, but if it's
going to cost 20% more, I'd rather not come back.  I believe passing on makes
most sense, but that should also condition coming to places that impose large

Randy: Thanks to the gentleman who brought in the remote participant from
Jabber, and that should be followed at all meetings

   8. IESG Open Mic

Olaf: I brought something up about proposed vs standards.  I'd like to thank
Jari for following up on the mailing list.  I don't want to discuss solutions
here, but I plan to write a draft.  Is the IETF list itself the best place to
discuss this?

Jari: I think a draft is the right process

Pete Resnick: Some of us have been talking about this in pragmatic ways.  For
example: we have a bunch of docs that could be full standard, how do we do
that?  Now, you just ask an AD, and they get last-called.  It has gotten easier.

Jari: We even talked about doing bulk approvals.

Andrew Sullivan: I have a modest proposal.  I was suggesting we should put
together... complicated joke :-)

Warren Kumari: another joke

x: <cannot follow statement>

Pat Thaler: Echo about trying to bring things that are stable to standards.  We
have in IEEE we tend to put our standards into ISO, and got pushback on
something because it references some RFCs that are at proposed standard, and
while we know they're not going to change on us, but ISO doesn't get it.

Ted: It might be nice to ask for advancement

Pat: We didn't imagine this would come up in advance, in this case we have to
deal with it without waiting.  But we'll look at it.

Ted: It's not a long process.  Even if it is too late.

It should be a 4-6 week process if the document does not require a revision.

Pat: We are trying to convince ISO that an RFC should be considered a standard,
the people who are objecting are not entirely neutral...

Olaf: I think it is important to recognise that the current language that
characterises proposed standards does not reflect reality and needs to change. 
I am working on a draft that reworks that characterisation, and will present it
for consideration.

We would like to see that happen

Keith: Given that this org has decided to go to two levels of standardisation,
perhaps someone should generate a list of old PS documents, and queue them all
for advancement consideration.  If something as immature as DKIM can be
advanced, we have already lost.

Barry: Some of the email documents need revision to advance, there are going to
be a lot of former draft and proposed standards

Sean: The RFC says we can uplift after two years

Barry: It's not that we can't, but we shouldn't without an update

Thomas: I think there's a perception problem here, it's not just advance the
standards.  If it's just a request and a button push, we've broken things.  We
need a better process for that

Ross: It seems we have a two step process, we should try to make it work. 
There may be PSes that are old and not widely deployed.  It's possible we may
have things that are out there that need revised, but I hope that most things
that have been stable for a while will be easy to advance.  I do intend to try
this out.  Once the obvious stuff has been advanced, there's going to be some
tough judgement as to when to stop.

Pete: there are a bunch of obvious cases.  Take consensus that they're what
should move forward, and that's it.