Hi Dave,

Since I see that John Leslie volunteered me to respond :-), I will give you
my thoughts so far - in a somewhat random order.  I am happy to brainstorm
and chat (or voice chat as well), if higher bandwidth is useful.

At IETF 97, I learned that Karen O'donoghue & Christian O'Flaherty are
working on the operationalization of IETF Hubs.  I have some discussion in
to have with them - but I don't think we've adequately connected yet, so
please understand that I probably have a different perspective and approach.

At IETF 95 in BA, the IESG did discuss having IETF hubs and was generally
in favor of encouraging them, via community/bottom-up organization.  We did
discuss trying to be proactive in clearing away organizational hurdles and
providing basic support.

As far as existing hubs go, I know that Christian has been involved in
Remote Hubs.  In addition, there is an IETF Hub in Bangalore (
[email protected]), but I don't see much traffic on that list and
an IETF Hub in Boston ([email protected]).  I also have a request
for a mailing list  for an IETF Hub in Northern Virginia, but I want to
connect fully with Karen & Christian before getting that set up.  There's
also an [email protected] for coordinating in-country activity
in India.

I really do believe that what is right depends very much on the geography
and how experienced in the IETF the interested folks are.

I would like to add some nuance to the description of IETF Hubs and
distinguish between the activity of being an IETF Remote Hub for one or
more WG meetings during an IETF meeting and being an IETF Community Hub.
A Community Hub is trying to build a semi-permanent community connected to
IETF through a variety of activities (social, technical talks, remote hubs,
etc.).

Let's talk briefly about what is necessary for a Remote Hub and some of the
associated challenges.

1) A physical location for the meeting.   Since an IETF Meeting has about 8
parallel sessions, trying to truly provide access to all of the sessions
would mean at least 8 different conference rooms.

2) Limited attendance and interest:  A particular geography is unlikely to
have folks interested in every WG - and perhaps not even more than 1.

3) Split Community:  Many active IETFers are off attending the physical
IETF meeting, which leaves those who can't go to attend the Remote Hub.
Sometimes that works fine - where there are folks who are well connected
into the IETF and can explain things as needed and yet don't attend all the
physical meetings - and sometimes it can leave an experience gap in the
room.

4) Inconvenience of time-zone: Very few people are interested in getting up
in the middle of the night to drive even 30 minutes to sit in a conference
room by themselves or with a couple other folks - and have the needed
participants for conversations be physically elsewhere.


Now, with all that said, my understanding is that having a Remote Hub with
at least one organizer that focuses on a small number of WGs for which
there is significant local interest can work well.  It combines best with
some social interaction (cookies equiv - going out to lunch/dinner/etc. )
and sometimes some introductory talks.

In the Boston area, Ravi Torvi from Juniper has organized Remote Hubs for
IETF 95 & 96 and is planning to again for IETF 98.  The time-zone
difference for IETF 97 was just too much.  The WGs covered tend to be
routing related, due to the interest of folks at Juniper.  Attendance has
varied, but rarely has more than a couple people who aren't usually at that
Juniper facility and has generally not had more than 10 people.  He has
tried having a social dinner one evening after a session and that has had a
bit more variety of attendance.  He hasn't tried having
additional/different talks.

In the Boston IETF Hub, we have and continue to be trying a few different
things.
   1) We have a couple organizers (Rich Salz and myself) and are looking
for more.
   2) Social interactions:  We've had a couple purely social get-togethers
- for dinner and to celebrate a retirement.   We are planning (but not
organized yet) that another social lunch or dinner should happen during
IETF 98.
   3) Technical talks: We have had two sessions of technical talks that are
related to IETF work and then followed by a social dinner for those
interested.  These seem to be quite successful based on feedback,
willingness of folks to present, and animated conversation.  Rich is
organizing the next session (planned for Feb 21) and it will be focused on
privacy and security.  We are trying this to see if having a focus topic is
popular.

Figuring out how to do effective outreach is and remains a challenge.  I am
sure that with help and focus that will improve and that's my primary hope
and concern at this point - beyond wanting another person or two to help
with the organization.

I realize that's a lot of background to give flavor and context to my
response to your questions.

On Sun, Jan 29, 2017 at 7:20 PM, Dave Crocker <[email protected]> wrote:

> Hi.
>
> For obvious reasons, there is some renewed/increased interest in use of
> remote hubs for IETF meetings.  Although there has been quite a bit of
> activity with hubs in recent years, I believe the expectations and
> requirements for them have been kept informal (and possibly idiosyncratic,
> with each hub doing whatever locals prefer.)
>

I strongly believe that what works for each geography and group of people
is different.  While I would like to see a slightly higher amount of
formality - like a well-known wiki that lists the various hubs and their
coordinators and gives each hub its own wiki page and mailing list if
desired - I don't see a lot of formality as yet appropriate.

If and as Hubs become more solid, then we have to think about
organizational considerations - such as the management of the coordinators,
how selecting, mentoring, and removing them works, and so on.  There are
also likely to be various IOAC-level considerations to think about.


> If we believe we are going to be relying on the use of remote hubs more,
> we should try to clarify what services they need to provide and how their
> operation should be incorporated into the conduct of face-to-face IETF
> meetings.
>

Personally, I think that we are relying on remote participation and
continuing to improve MeetEcho is great and useful.  I'd  recommend taking
a quick read through
https://datatracker.ietf.org/doc/draft-manycouches-completely-virtual-meetings/
as
well.   I just don't think that we will have sufficient density of
participants in any particular WG in any particular geography to have the
meaningful conversations not take place via MeetEcho - and I'm not very
comfortable with a split conversation anyway.

There are already a number of interactions that WG Chairs have become
accustomed to for MeetEcho and I think that they mostly generalize
regardless of whether there is one person/remote room or more than one.


> By way of example...
>
>    Should they support multiple, parallel sessions, so that different
> groups of hub participants can 'attend' different, parallel sessions?
>
>    Should the interaction between hub participants and participants at the
> 'main' venue be subject to particular procedures/style?
>

We've talked repeatedly about having a virtual queue that everyone gets
into - local and remote, but there are concerns around impacting the speed
and interactivity of the in-person interactions too.  Let me turn it around
- what do you see as being needed beyond what we had at IETF 97?

Regards & thanks for reading this far too long email,
Alia


>
> Thoughts?
>
> d/
> --
>
>   Dave Crocker
>   Brandenburg InternetWorking
>   bbiw.net
>
> _______________________________________________
> NOTE WELL: This list operates according to
> http://mipassoc.org/dkim/ietf-list-rules.html.
> https://www.ietf.org/mailman/listinfo/vmeet
>
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html.
https://www.ietf.org/mailman/listinfo/vmeet

Reply via email to