Re: procedural question with remote participation


Ugh.  Ignore that email below - I had sent it a few days ago but somehow it got 
stuck in the outbox and never got sent, and the discussion is past that point 
now so it doesn't matter.

-hadriel


On Aug 12, 2013, at 12:35 PM, Hadriel Kaplan 
<hadriel(_dot_)kaplan(_at_)oracle(_dot_)com> wrote:


[personal disclaimer: I have participated remotely, a few times, and I agree 
that it's not the same as being there, and I agree that it could be improved. 
 But I think we need to balance the needs of remote participants, vs. the 
goals of physical meetings: to get work done that can only get done with 
direct human interaction, with real-time interactive discussion, and hallway 
co-mingling, etc.  In my opinion being remote will *never* be the same as 
being there, and that's why we have physical meetings to begin with, instead 
of just virtual ones - ultimately I want you to come to the meeting 
*physically*... but I want to do as much as possible and *practical* to 
accommodate remote participants, because I know not everyone has the luxury 
to come, and I want their input regardless.]


I think one challenge with this discussion, at least for me, is that it feels 
like the solutions being proposed are completely out of whack with the 
severity and nature of the problems at hand, and the people they apply to.  
Adding more rules and processes is about the surest way I know to annoy 
engineers, let alone volunteer ones.  And if your right hand is sore, you 
don't cut it off and replace with a mechanical one, or go build a 
billion-dollar robot to be controlled by brain-waves.

People on this list are saying things like "we don't know who's at the mic", 
and "we can't see the slides well enough", and "presenters are hard to 
follow".  ISTM there are practical and low-tech means of fixing those 
problems.

Here're some ideas of possible solutions - but most of these boil down to 
either "have WG Chairs do their job", or "get over it":

Problem-1: we don't know who's at the mic, because people forget to say their 
names.
Solution: (a) train WG Chairs to remind people at mic, and interrupt them if 
they haven't said their names, (b) have jabber scribes sit next to mics, to 
do the same as WG chairs, (c) if chairs and scribes forget, send a chat 
message in jabber to remind them.  Is it annoying?  Sure.  Will it sometimes 
fail to work?  Sure.  But every other solution will also be annoying and not 
always work, and this solution is very low-tech and simple.  This is what 
happens in RAI area WG meetings, and appears to work afaict.  If it doesn't 
work, then we go to a Plan-B in the future.

Problem-2: we can't see the slides well enough, in video.  Meetecho is ok, 
but they don't cover all meetings.
Solution: we pay Meetecho to cover all WG meetings.  Last time I checked, we 
weren't paying them anything; but if we really want that type of technology 
in all meetings, then it's only fair to pay them. (frankly we should be 
paying them now already)  We also need to remind people to use reasonable 
font size.  10pt is bad mojo even in the physical room.  That type of thing 
is what a WG chair should be telling people, though - we don't need slide 
police.

Problem-3: presenters are hard to hear/follow/understand.
Solution: that's life.  We're a volunteer organization, not trained 
professional thespians.  We have drafts written in plain ascii available in 
advance, and audio, and video, and jabber, and (hopefully) Meetecho-type 
service, and that's about as good as it can reasonably be.

Problem-4: we want slides 7 days in advance to translate to our native 
language
Solution: You're doing it wrong - or rather, the presenters are doing it 
wrong.  The slides aren't a replacement for drafts, or a copy-paste of draft 
text.  WG Chairs know this, and should remind presenters of it.  As an aside, 
if you really need a week to translate them, it's highly unlikely you'll be 
able to truly *participate* interactively during the meeting - and you don't 
need to be real-time if all you're going to do is listen/watch - we record 
the whole thing for your later viewing pleasure.

Problem-5: we want slides 48 hours in advance to translate to our native 
language, download it, etc.
Solution: sounds reasonable.  Have WG chairs use 48 hours as their guideline, 
but not as a strictly enforced rule a la I-D submission deadlines.  WG chairs 
can figure out when exceptions need to be made, how to deal with it, etc.  We 
don't need this in an RFC.

Problem-6: if you change slides, I don't know which slides you changed
Solution: it happens, and that's life.  You've got to follow the speaker's 
words, not a published script.  The point of these meetings is for real-time 
discussion on relevant matters that can only be handled live.

Problem-7: 

-hadriel


On Aug 8, 2013, at 10:43 AM, John C Klensin <john-ietf(_at_)jck(_dot_)com> 
wrote:

--On Tuesday, August 06, 2013 11:06 -0400 Andrew Feren
<andrewf(_at_)plixer(_dot_)com> wrote:

...
I think this sort of misses the point.  At least for me as a
remote participant.

I'm not interested in arguing about whether slides are good or
bad. I am interested in following (and being involved) in the
WG meeting.  When there are slides I want to be able to see
them clearly from my remote location.  Having them integrated
with Meetecho works fine.  Having slides and other materials
...


Let me say part of this differently, with the understanding I
may be more fussy (or older and less tolerant) than Andrew is...

If the IETF is going to claim that remote participation (rather
than remote passive listening/ observation with mailing list
follow up) is feasible, then it has to work.  If, as a remote
participant, I could be guaranteed zero-delay transmission and
receipt of audio and visual materials (including high enough
resolution of slides to be able to read all of them) and that
speakers (in front of the room and at the mic) would identify
themselves clearly and then speak clearly and at reasonable
speed, enunciating every word, I wouldn't care whether slides
were posted in advance or not.  

Realistically, that doesn't happen.  In some cases (e.g.,
lag-free audio) it is beyond the state of the art or a serious
technical challenge (e.g., video that is high enough resolution
that I can slides that have been prepared with 12 point type).
In others, we haven't done nearly enough speaker training or it
hasn't been effective (e.g., people mumbling, speaking very
quickly, swallowing words, or wandering out of microphone or
camera range).   And sometimes there are just problems (e.g.,
intermittent audio or video, servers crashing, noisy audio
cables or other audio or video problems in the room).   

In those cases, as a remote participant, I need all the help I
can get.  I'd rather than no one ever use a slide that has
information on it in a type size that would be smaller than 20
pt on A4 paper.  But 14 pt and even 12 pt happen, especially if
the slides were prepared with a tool that quietly shrinks things
to fit in the image area.  If I'm in the room and such a slide
is projected, I can walk to the front to see if if I'm not
already in front and can't deduce what I need from context.  If
I'm remote and have such a slide in advance, I can zoom in on it
or otherwise get to the information I need (assuming high enough
resolution).  If I'm remote and reading the slide off video,
especially low resolution video, is hopeless.  

More generally, being able to see an outline of what the speaker
is talking about is of huge help when the audio isn't completely
clear.  Others have mentioned this, but, if I couldn't read and
understand slides in English easily in real time, it would be of
even more help if I had the slides far enough in advance to be
able to read through them at my own pace before the WG session
and even make notes abut what they are about in my most-familiar
language ... and that is true whether I'm remote or in the room.

And, yes, for my purposes, 48 hours ahead of the WG meeting
would be plenty.  But I can read and understand English in real
time.  If the IETF cares about diversity as well as about remote
participation and someone whose English is worse than mine is
trying to follow several WGs, 48 hours may not be enough without
requiring a lot of extra effort.

That is not, however, the key reason I said "a week".  The more
important part of the reason is that a one-week cutoff gives the
WG Chair (or IETF or IAB Chairs for the plenaries) the time to
make adjustments.  If there is a nominal one week deadline, then
the WG Chair has lots of warning when things don't show up.  She
can respond by getting on someone's case, by accepting a firm
promise and a closer deadline, by finding someone else to take
charge of the presentation or discussion-leading, or by
rearranging the agenda.  And exceptions can be explained to the
WG on the mailing list.  With a 48 hour deadline, reasonable
ways to compensate are much less likely, the Chair is likely to
have only the choice that was presented this time (accepting
late slides or hurting the WG's ability to consider important
issues) and one needs to start talking about sanctions for bad
behavior.   I would never suggest a firm "one week or no agenda
time" rule.  I am suggesting something much more like a "one
week or the WG Chair needs to make an exception, explain it to
the WG, and be accountable if the late slides cause too much of
a problem".  There is some similarity between this and the
current I-D cutoff rule and its provision for AD-authorized
exceptions.  That similarity is intentional.


--On Monday, August 05, 2013 13:36 -0500 James Polk
<jmpolk(_at_)cisco(_dot_)com> wrote:

At 12:38 PM 8/5/2013, John C Klensin wrote:

Hi.

I seem to have missed a lot of traffic since getting a few
responses yesterday.  I think the reasons why slides should be
available well in advance of the meeting have been covered
well by others.  And, as others have suggested, I'm willing
to see updates to those slides if things change in the hours
leading up to the meeting, but strongly prefer that those
updates come as new alides with update-type "numbers" or other
identification rather than new decks.  In other words, if a
deck is posted in advance with four slides numbered 1, 2, 3,
and 4, and additional information is needed for 3, I'd prefer
to see the updated deck consist of slides 1, 2, 3, 3a, 3b, 4
or 1, 2, 3a, 3b, 3c, 4, rather than 1, 2, 3, 4, 5, 6.


How exactly do you do this in pptx? Numbering slides is a
linear operation AFAICT, and it's binary (it's either on or
off). Please educate me if I'm wrong; lord knows I don't know
don't know how to do everything flag/setting in powerpoint...


Hmm.  First, I missed the IESG Statement requiring the people
use pptx or any other particular piece of
presentation-preparation software, much less the IETF consensus
for that statement.  If the tools you choose to use don't meet
the needs of the IETF, that doesn't seem to me to be an IETF
problem.

Second and more important, while having slide preparation
software number pages, I'm not aware of anything that prevents
you from typing in numbers, especially with something that is as
inherently page (slide)-oriented as presentation-preparation
software.  Yes, it is an annoyance, but having the need to
update slides within a short interval before the meeting/
presentation/ discussion is an annoyance to participants who
depend on the ability to access those slides and who might want
to prepare reference or discussion notes in advance.

And, in my 8 years as TSVWG chair, I've rarely had completely
new individual slides sprinkled throughout an existing deck.
Rather, I've received updated slides - each with part of their
content altered. Does this fall into your desire for a "3a",
or is that just "3" (because 3a means an entirely new slide
from scratch)?

3a.

BTW - I'm very much *not* in favor of stipulating to my WG
that slides must be turned in 7 days in advance of a TSVWG
meeting. I personally think no more than a 48 hour advanced
window should ever be considered.


See above.

 best,
 john