Re: [103attendees] A/V in Bangkok

[Brian E Carpenter <[e-mail]>]

On 2018-11-10 09:29, John C Klensin wrote:
> 
> 
> --On Friday, November 9, 2018 15:06 -0500 Ted Lemon
> <[e-mail]> wrote:
> 
>> There were a lot of cases, particularly early on, when remote
>> presenters' audio clipped badly, making it painful to sit in
>> the room.   That seemed to improve over the course of the
>> week, but the gain on remote presenters in every single remote
>> presentation I heard was still too high, again making it
>> painful to sit in the room.
>>
>> Quite a few times when someone wanted to comment remotely they
>> couldn't, because their microphone wasn't working.   This
>> wasn't necessarily meetecho's fault in a direct way, but we
>> have to figure out how to get this right if we're serious
>> about remote participation.
> 
> Ted, 
> 
> Very helpful.  Thanks.  Wrt the remote presenters and borrowing
> from some earlier comments from one of the Meetech team, is
> there any way to differentiate how many of those problems with
> remote presenters of whom Meetecho was notified well in advance
> and with whom they have an opportunity to test and debug (as
> they have been requesting for several years)

At least in my case, I'd done the test, discovered that my browser
(Waterfox) couldn't use any of the STUN/ICE options, and switched
to a browser (Firefox) that passed the tests. But when I came to
put my hand up in a WG, the floor control for some reason failed
to enable me, so I couldn't speak. Fortunately there was an active
jabber scribe in that session who read out my comments.

(Before anyone points out that Waterfox and Firefox are virtually
identical: yes, I know, and all the relevant config options were
set identically. But something was different. Just one of RTCweb's
little mysteries.)

I wonder if it would be possible to setup a dummy WG session during
the last few days before each IETF, where anybody could do a test,
with some non-meetecho person acting as chair, and meetecho
watching as they do for real sessions.

> and those who
> simply used the remote queue facilities as if they were just
> making a comment?

afaik, the meetecho test is supposed to validate both that mode and
presenter mode. But the devil is in the details.

   Brian

> 
> thanks,
>    john
>