(directed to <vmeet> only:) Peter Saint-Andre <stpe...@stpeter.im> wrote: > On 3/29/12 2:50 AM, George, Wes wrote: >> ... On Behalf Of Melinda Shore >>> >>> I've put up a first crack at a how-to-do-remote-good page, here: >>> >>> http://wiki.tools.ietf.org/group/wgchairs/wiki/RemoteParticipation. >> >> I made a few changes in the wiki page... > > This is a very helpful page. I made a few tweaks, but not much was needed.
It's a very helpful page for remote-attendance. :^) It's, alas, not up to the task for remote _participation_. :^( "Typically 5-15 seconds" isn't good enough. (I'm not arguing whether _one_ second is even "good enough".) The delay needs to be known. Adjusting to a five-second delay is one thing; adjusting to a 30-second delay (which is too-often seen) is quite another. Adjusting to an unknown delay is no longer "participation". When asking whether the remote audio is OK, _measure_ the delay. The part about one person being both jabber-scribe and minute-taker needs to be "Don't do this!" If remote participation is important, minutes need to happen separately. To some degree, minutes can be pieced together after the fact, provided there is a separate backup audio recording. The Jabber Scribe needs full-time access to a microphone. Arguably the Jabber Scribe should be sitting next to the meeting Chair. Mentioning the issue of adjusting for the audio delay when recognizing remote participation is good -- but there needs to be a practical way of accomplishing this. Sitting next to the meeting Chair and mentioning who wants to be channeled seems plausible... Projecting the jabber stream probably isn't a great idea for the general case (there will certainly be exceptional cases). Advising the in-room participants to use the jabber room is always good, though. The part about clarifying the name of each speaker at the microphones is a start, but again there needs to be a practical way... In practice, what I find works best is "Who's talking" questions in jabber, followed by "Please state your name for the minutes" if nobody answers in jabber. (Of course, this coming 30 seconds _after_ the person mumbled his/her name _is_ disruptive, but IMHO it's important enough to justify the disruption.) It's asking _a_lot_ of the Jabber Scribe to note the name of every person who speaks, but possibly the Jabber Scribe could guess and add a question-mark when s/he's at all uncertain. (If the Jabber Scribe already has no idea whatsoever, asking on-mike is called-for IMHO.) I suggest that the jabber stream SHOULD contain the name of every participant speaking at the microphone (or channeled). In practice, the Jabber Scribe can't simultaneously type and speak; so someone else would probably need to add the "channeling NN" note. <asbestos-suit = ON> -- John Leslie <j...@jlc.net> _______________________________________________ NOTE WELL: This list operates according to http://mipassoc.org/dkim/ietf-list-rules.html. https://www.ietf.org/mailman/listinfo/vmeet
