Several comments from the perspective of someone who is remote at most IETF meetings these days but often has a lot of comments and questions.... --On Thursday, November 8, 2018 08:29 +0100 Mikael Abrahamsson <swmike@xxxxxxxxx> wrote: > I just use the Meetecho recording page. Yes, I need to watch > the stream even if I do not need it (because I am in the > physical room), but I do get a combined view of chat room, > live video feed, and slides in my browser without needing to > install any other software or having an account anywhere. This certainly works. However, at least for those of us with significant available screen real estate when working remotely, it has two disadvantages. (i) If one does want to watch the panel with the slides and the speaker stream, there is a certain amount of nuisance switching back and forth required that makes it hard to keep track of the chat activity as well. It is much easier if the latter is in a completely separate window that can be positioned as needed. But that requires a Jabber client that is not embedded in a web browser, much less in Meetecho, unless one wants to run two separate browser windows. (ii) The input window is tiny (one line and narrow), making typing a question or comment more complicated than "well, the room isn't too cold where I am" or "Joe Bloggs at mic" and proofreading it without sending much more complex and time-consuming than it should be. I have given up: I compose non-trivial remarks and questions in a text editor, check them, and then paste them into Meetecho and send them. Still not optimal. Neither of those issues is particularly relevant to someone monitoring Jabber in the meeting room as distinct from (a subset of) those of us was are working remotely. (The Meetecho team knows how I feel about these things and I understand why fixing them within Meetecho is probably not the right solution.) > When I am a jabber scribe, I type into the chat room the name > of the person speaking at the mic as people walk up to the mic > (I read their badge (PLEASE PEOPLE, HAVE YOUR BADGES EASILY > READIBLE!), and this means together with the audio, video and > slides, people speaking into the mic are identified by name > and the note taker can use this to make sure the notes are > correct). Dan and several others do this as well. It may not have been said publicly before, but it is hugely helpful and not just for note takers. For those of us who are remote, being able to know who is speaking is very important -- it provides context for those of us who have been around for a while and may help newcomers (in person as well as remote) get to know the community. While people are supposed to announce their names, many forget and even more, who know their own names and are anxious to get on with their comments, tend to go to the microphone and quickly mumble something entirely unintelligible to those listening remotely and trying to simultaneously listen to (not always perfect) audio and track Jabber and the slides. Video rarely helps much either, as, in most WGs, the Meetecho camera almost always stays pointed at the front of the room (main microphone or chair's table). (Not the fault of the Meetecho team -- there are no solutions that that problem which generalize and are optimal at least unless we are willing to go to the considerable additional cost --in people, rather than technology-- of running two cameras and possibly having both feed to the client screen). So, much of the time, "XYZ at mic" is our only real clue about who is speaking. > I also start each session by stating that I am the jabber > scribe and that people who want things relayed to the mic > should preface their statement with "to mic:". This has worked > well for me and the sessions I jabber scribe. Oddly, it has also been a source of confusion. <small diversion> It seems to have not been publicized as well as the Meetecho team and apparently the IESG thought it was, but there are now two ways for a remote participant to inject a comment or participate in a discussion. One, as we know, is to go through Jabber and use "to mic:" or "MIC:". The other is to use Meetecho's "remote queue" facility, which actually gets the remote participant into a queue analogous the the microphone line(s) (Anyone who is reading this who doesn't know, look for the little hand next to your name above the participant panel.) WHen it is feasible to use, it has at least three advantages over typing a comment into Jabber: (i) It doesn't require typing before making the request to be heard or then waiting for someone to get in the back of the mic line, which lowers the odds that the discussion will have moved on to something else before the comment is heard. (ii) If one is actually trying to participate in a discussion, being "live" on audio and video allows interactions much more like standing at a floor microphone and using a human relay for typed text. And (iii) Meetecho notifies the WG Chairs directly (and apparently makes noise, rather than requiring active monitoring) when there is someone in queue. Comments from IESG members (and especially Alissa) seem to imply that the remote queue mechanism is not only their preferred choice but that they think most people have switched to it rather than using Jabber. The bad news is that, at least from my experience, the facility is still in need of considerable tuning, not in how Meetecho works but in how WG Chairs and equivalent handle it. More training needed and all that. At the same time, I think we need to keep Jabber and keep it working for remote input until we discover an equivalent solution. The remote queue arrangements don't work for anyone who is in a location where their talking to the the computer or running video input would be disruptive or violate local rules. Sending audio and video from the client requires significantly more high-quality bandwidth than simply receiving Meetecho and occasionally typing a few lines into Jabber, and some IETF participants may not have that all of the time (or ever). People who feel uncertain about their English may be more comfortable typing and proofreading than speaking (although my preference would be to convince them that we are less fierce than we often appear and that, if they speak slowly and clearly, we will happily adjust). And some comments may be brief and self-contained enough that opening up an A/V feed from the client would be rather high overhead. There may be other reasons in particular WGs or other sessions, including the observation that Jabber input makes a record of the question even if it is not repeated at room microphone while a request to get in the remote queue leaves few tracks if the chair doesn't turn microphone of the person making the request on. </small diversion> When I've seen something equivalent to "if you want something repeated in the room, start it with 'to mic:'", I take it to mean "Jabber input is preferred in this session to use of remote queues". Maybe that will go away over time as we all get used to the remote queue facility but, for the near term, I'd much prefer to see something more like "if you have input for the room and don't find the remote queue facility convenient, type your message into Jabber and preface it with 'MIC:'" > I also recommend that the chairs in sessions monitor the > jabber room. Some do already, but I recommend everybody to do > so. Good idea but, especially for very controversial discussions and plenaries, may not be realistic. best, john
