WG Meeting Structure / Alternate Room Layout

[Adam Roach <[e-mail]>]

Hi. I'm following up from a recent thread on this mailing list ("WG meeting
structure" [1]) discussing the prospect of running additional meetings 
with a
"U-Shaped" table structure for the purpose of facilitating more natural
back-and-forth discussion of topics.

As many of you will probably recall, we ran an experiment at IETF 98 in 
Chicago
in which several working groups opted in to using such a layout. Those 
groups
that participated were QUIC, REGEXT, NFV, LAMPS, and HTTPBIS.

The stated goal of this experiment was basically the same as the goal put
forth in the thread cited above: to reduce the formality of a strict
microphone queue and encourage a more free-flowing back-and-forth 
exchange, in
the hope that this more informal style of discussion would lead to more 
rapid
progress.

The feedback that the IESG received immediately following the experiment was
mixed, with significantly more negative feedback than positive.


An exhaustive list of the positive feedback received follows:

   1.  A QUIC chair indicated: “seems to have worked well.”

   2.  One QUIC attendee noted that it “allowed a much greater
       volume of comments to be made by cutting down on overhead.”

   3.  Remote participants were able to more easily see expressions/
       reactions of people at the table


The negative feedback included the following points:

   1.  It is hard to tell who is speaking (almost everyone who gave
       feedback mentioned this)

   2.  Chairs found that sitting with their back to room was sub-optimal
       (almost all chairs indicated this). This had two reported impacts:

       –  The chairs can’t see people queuing at the floor mic

       –  The chairs don’t get a good feel for room assent/dissent

   3.  Single mic for non-table participants required long walks for
       some people (several participants indicated this)

   4.  Division between table and non-table perceived as elitist,
       especially when off-mic side-conversations broke out at the
       table (2 comments)

   5.  Having to turn head 90° to see the slides is uncomfortable
       (2 comments)

   6.  Chairs had no dedicated space, no dedicated mic, were pulled
       aside by neighboring IETFers into side conversations during
       meetings.

   7.  Several people who elected to sit at the table were not active
       participants, and took up space that was needed by active
       participants. Attempts to relocate these people were awkward
       and occasionally ineffective.

   8.  People were reluctant to fill in on “side” seating area, opting
       instead to stand at the back of the room.


While the volume of negative feedback was significantly greater than the
volume of positive feedback, there are two factors that should also be
taken into consideration. The first is that people are typically more
proactive about giving negative feedback than positive, so actual sentiment
on the topic may skew more positive than the preceding summary indicates.
The second is that several of the issues described in the "negative 
feedback"
list are not inherent in the revised layout, and may be possible to 
address if
the experiment were re-run.

Unfortunately, while there were plans to solicit feedback on this room 
layout
as part of the post-meeting survey for IETF 98, due to a technical 
issue, that
survey was never distributed to the IETF community.

In the more recent thread on this WG Chairs mailing list, the following 
issues
were either re-iterated from the initial feedback above, or mentioned by 
more
than one person:

     - Remote participants found meetings in the alternate layout more
       difficult to follow.

     - Many people who could not fit at the front tables felt excluded.

     - Everyone (both local and remote) had difficulty figuring out who was
       speaking, including Jabber scribes.

     - The style of queuing discipline required for this layout is
       unfamiliar to existing WG chairs, and many did a poor job with it.
       Non-front-row and remote participants particularly suffer from this.

     - There was no direction for chairs to face where they could see
       everything that was going on.

At the same time, there were approximately as many comments that such an
alternate layout seemed like a good idea to try. It was unclear in some 
cases
whether commenters were aware of the previous experiment or the community's
comments on it.

There were several suggestions for incremental improvements on the previous
experiment, both in the post-IETF-98 feedback and during the more recent
discussion on the WG Chairs list. Some of these included:

     * Place chairs where they can see the microphone

     * Give chairs dedicated table/mic

     * Experiment with shape of “U”
         – Face toward room instead of away from it
         – Make into open “V” facing room

     * Use two or more screens to display slides

     * Only allow this layout for very small groups

     * Use more and/or a different style of microphones


Given the foregoing, including in particular those aspects of negative
feedback that are inherent to the layout, the IESG would like feedback 
from the
chairs regarding whether a re-run of the experiment might be worthwhile.

/a
____
[1] 
https://mailarchive.ietf.org/arch/msg/wgchairs/rMPdilySo8XNU84623TQt5lRnJY