Re: [ga] Draft of Bylaws Recommendation concerning the General Assembly

["Ben Edelman" <edelman@law.harvard.edu>]

(Note that I have received Kendall's 2/13 comments.  Am thinking about them,
and about possible adjustments that might respond to his suggestions.  Hope
to provide a response within the next few days.)

In Joanna's last message, she made, to my eye, four major suggestions:

1) Displaying the chat room on screen.

2) Increased fairness and even-handedness in selecting remote comments.

3) Accepting remote comments in advance of the start of the meeting.

4) Increased reliance on multimedia participation, perhaps via 6-10 folks
selected to participate in some special way for the occasion.



I'd like to respond to each suggestion in turn.



1) Re displaying the chat room on screen:

First, I hadn't previously understood that you were proposing to show the
chat room, rather than the remotely-submitted text comments (which, as you
may recall, are in a separate system -- web form submissions of longer
length and easier editing than is usually characteristic of a chat client).
Accordingly, you'll notice that my last message speaks only to the question
of displaying submitted text on screen, not the chat room specifically.

Now that I better understand your suggestion, some further responses:

* My "distraction" comment from the prior message stands, and I stand by it
even more forcefully than I did in the prior instance, even as I understand
that you are not convinced by this reasoning.  Again, based on prior
experience, when I did this in multiple prior meetings, I believe this is
distracting to at least some attendees.  There's just something different
about an in-person meeting (where the usual norm is to concentrate on the
one person talking) versus a webcast via computer (where the medium seems to
call out for multiple means of participation).

* The chat room really isn't intended to be a place of substantive,
on-the-record deliberation.  The introduction text viewed upon entrance to
the chat room says so pretty clearly.  And much of the discussion in the
chat rooms, in my experience (and as the logs reflect), is quite
nonsubstasntive.  I'm not anxious to project chat room comments re audio
problems, login problems, and the other various glitches that some folks
experience at various points in the meeting; it just doesn't seem helpful to
the work at hand.

* I'm concerned that the stream-of-consciousness tone of the chat room isn't
appropriate in the setting of the in-person public meeting.  Furthermore,
general and enduring display of such messages might cause the in-person
participants not to take remote participants seriously.  Instead, I think
remote participation might be better received if remote participants
contributed occasional well-thought-out comments, rather than continuing
marginally-on-topic banter, for lack of a better word, which truly has been
my experience in prior meetings.  (That is, in my experience, the chat room
comments often stray from the discussion in the in-person meeting, and I'm
not sure that's helpful on balance.)  Ultimately, the chat room just doesn't
seem consistent to me with the tone of a public meeting, with the tone of
the speakers at the microphone, and with the expectations of those attending
the meetings in person.

* On some occasions, we've had some pretty serious violations of chat room
policy -- personal attacks, flames, obscene comments, etc.  I've never had
to ask anyone to leave the chat room, nor to remove anyone using the
technical means at my disposal, but I've come close (issuing "this is your
second and final warning" messages).  Again, this sort of thing isn't
appropriate for display to the general audience, and neither would it be
helpful in that context.  It's all fully & publicly archived if anyone is
interested, but projecting it just doesn't make sense to me.



2) Re increased fairness and even-handedness in selecting remote comments.

Your comments here, as well as Kendall's, raise a serious set of questions,
to which I hope to reply at some length, pending further discussion among my
team, as well as other personal obligations (other work to be done!) and
some further investigation re our technical capabilities.  That said, here's
what immediately comes to mind.

You wrote:

> The principle is that remote participation should not be manipulated
> unnecessarily as this gives false feedback to those in the public
auditorium
> about the nature and extent of online commentary.

Certainly I don't think of the current selection process as "manipulation,"
so I think some commentary is in order here.

When in November 1998 I first wrote the code that is the basis of the remote
participation system still in place today, my goal was simple: to take
public comments, in real-time, in a way that hadn't been possible at the
IFWP meetings that preceded ICANN's formation.  It seemed immediately clear
to me that we'd get more online comments than we'd have time to read, and
indeed this proved true in every ICANN meeting.  But simply reading the
first few comments was clearly an unacceptable methodology; after all, that
would encourage a race to submit comments as early as possible, clearly (I
thought) an undesirable outcome.

In response to concern about how to select questions to answer, two
significant systems were put in place: First, a "meeting section" selection
box -- in which users would self-classify their comments to map to agenda
items on a drop-down list.  The "ask a question" screen clearly reports
which section is currently in progress, and the Remote Participation Liaison
chooses comments for a particular section's Audience Q&A session exclusively
from the comments marked for that section.  Second, a Remote Participation
Liaison was carefully selected and trained -- in the first instance by
Jonathan Zittrain, the Berkman Center faculty member working on the Center's
initial remote participation efforts -- to select comments solely for their
clarity and relevance to current discussion, but not for a particular
substantive position, one way or the other.  That's not to say that our
efforts on that front have been successful in every instance -- I imagine
there are plenty of allegations that they have not! -- but it was certainly
our goal, I hope pretty clearly articulated in the documentation &
instructions for the remote participation system.

Kendall suggested an interesting alternative way to select comments -- some
variation of a vote by online participants.  I'm still investigating the
feasibility & practicality of doing so, as well as other concerns like fraud
control, perceived fairness, etc.  Again, a response on all this is
forthcoming.



3) Accepting remote comments in advance of the start of the meeting.

You wrote:

> ... I would think about requesting some questions and comments through
> the GA list, submitted in advance.

We have considered this in the past, but have ultimately concluded that it's
not the best way to proceed.  ICANN and the DNSO maintain numerous methods
of receiving comments from online participants in non-real-time contexts --
public mailing lists (including this one), the ICANN public comment boards,
messages sent directly to staff, Board Members, NC members, constituency
representatives, etc.  It's certainly not the intention of the webcast
efforts to replace or compete with any of these systems, and we don't seek
to do so.  Accordingly, our rule, since the November 1998 meetings, has been
to accept only real-time (not presubmitted) comments.  In practice we'll
accept any comment meeting the length requirement (less than 1000
characters -- roughly the amount we can display on a single projection
screen in a large font) and have no way of knowing when a remote participant
actually drafted the comment.  But the responsibility is on the remote
participant to submit a comment during the meeting if it is to be treated as
a comment submitted to the meeting itself, as distinguished from the
non-realtime comment-receiving systems.

For whatever it's worth, this is (as best I can remember) the first time
I've received this comment, i.e. the suggestion that presubmitted comments
should be accepted along with "live remote participant" comments.  My sense
previously was that everyone thought it was a good idea to require that
comments be submitted during the meetings; no one has ever said otherwise.

> If not, you have the task of selecting
> questions on the fly, while trying to ensure a fair and even handed input,
> without knowing what's going to hit the screen next, or when you're going
to
> get a decent comment to read out.  This is not a trivial matter.

I'm a little confused about what you mean here.  I think you may be
confusing the live chat room (an "IRC chat," like an AOL chat room or some
such) with the remote comment submission system (web forms that submit text
to a database).  Our remote participation liaison can readily review all
messages in the database, can mark messages for (one-button) display on the
projection screen, can classify messages according to a variety of criteria
(whether or not read yet, meeting section if not so classified by the
submitter, etc.), and can generally keep pretty good track of what's going
on.  I'd like to think that we do know quite well what's going to hit the
screen next!  (Indeed, each comment is displayed on-screen while it's
read -- as the scribe's notes generally reflect, i.e. "remote comment #1234
from Tom Smith read in its entirety and displayed on screen" might be
entered into scribe's notes when that is the case).

In terms of fairness and evenhandedness, I understand the concern here, and
agree that it's a serious one.  But please understand the other issues we
also face here: For example, I hope you appropriately consider the concern I
expressed in my last message about reading & presenting off-topic messages
(perhaps "stale" -- read too late, i.e. after the topic in question was
discussed by the meeting -- or perhaps inherently off-topic).  Doing so
causes the audience to literally cringe, in my experience, as if to say "Why
are those Berkman Center people so dumb as to read us a question about
*that*, when we just finished discussing it, and were all quite tired of the
topic after such a prolonged discussion of the subject?"  For this reason,
and for others too (ad hominem attacks, questions not appropriately
addressed to the current meeting or group, messages submitted from what
immediately seem to be false identities, etc.), we've come to think that any
automated way of selecting comments is likely to be undesirable for other
reasons, notwithstanding the underlying desirability of an automated way of
choosing messages to be read aloud.  Again, I look forward to responding to
Kendall's suggestion on this subject, and so I'd propose to bracket this
discussion for the moment.


> We could aim for consensus in the GA as to the topics that need
highlighting
> in this way. Questions could be supplemented or even substituted by live
> comments on the day if desirable, but it would give the online
participation
> some much needed structure and ensure quality questions are heard on a
wide
> range of topics at agreed points in the schedule, interwoven with those
from
> the audience (some of which I'm sure are prepared in advance).  I'd be
very
> surprised if a call for questions did not receive a very positive response
> from the GA list over say a 2 week period.

Certainly any effort to increase GA participation both in the GA's physical
meeting and in other ICANN sessions seems to me highly desirable, and I
applaud such attempts.

If these attempts are successful and lead to insightful and clearly-stated
questions closely related to the topics under discussion on the various
meeting agendas, I'm confident that the remote participation liaison will
properly present these messages to the assembled group.

My sense in prior meetings has been that, in general, the liaison is usually
successful in presenting a reasonable range of messages, giving appropriate
recognition to major topic areas of the submitted questions and reading a
substantial portion of the messages received.  I know that there have been
instances when this has not been the case -- when public comment (both
in-person and via remote participation) has been cut short due to shortness
of time, or when the meeting chair has used his or her prerogative to
recognize disproportionately many in-person commenters.  (Incidentally, I do
continue to believe remains a significant factor to keep in mind -- the
assembled group, and the queue at the microphone, can prove quite a force to
be reckoned with, given their physical presence, travel expenses incurred,
and general immediacy, especially as against a seemingly invisible and
distant online audience.  I continue to suggest to the chair of each meeting
that a remote comment be recognized after every two microphone comments, and
at times we've gotten into an excellent rhythm this way.  But the underlying
issue surely remains.)



4) Increased reliance on multimedia participation, perhaps via 6-10 folks
selected to participate in some special way for the occasion.

You wrote:

> ... Also, two
> contributors are not sufficient in number to establish this format as part
> of the process, hence it is easily reduced to an irritant in the
> proceedings. My suggestion would be to have 6-10 contributors, as the
> schedule allows, either interwoven at regular intervals or introduced as a
> block, but all using the same screen, speaking individually in rotation.

Can you propose a specific technical means (hardware and software) for doing
so?  At the moment I'm a bit unclear on what specifically you're talking
about -- videoconferencing, or something else.

How would we choose the 6-10 contributors?  No one gets a similarly-favored
position in the physical meeting, so I'd be concerned that this approach
might not receive the support of physical meeting attendees.



Again, I thank everyone on this list for their suggestions.  I remain
interested in further comments -- everything from improving the wording of
web pages announcement emails to more far-reaching comments like the ideas
submitted so far.  I certainly can't promise that the Berkman Center will do
everything that's suggested here -- but we're listening, and, we think,
thinking seriously about these issues.



Ben Edelman
Berkman Center

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html