To help us get on track for our next show (Expo SF, May 3 – 6), I worked with the team to draft a new plan. Among the things we’re going to do: 1) no tweetstreams displayed publicly in the presentation rooms (unless a presenter wants it); 2) for the keynotes, we’ll monitor the tweets and give any appropriate feedback to the presenters who want it (although live-tweeting and commentary are common, feedback is actually pretty rare, so we anticipate this will happen maybe twice during the whole show); 3) we’ll work with the AV team to figure out a way we can convey a message to presenters during their talks (again, only if they want it); 4) we’ll work with presenters before they go on stage to prep them on ways they can note and respond to feedback.

We’ve got more ideas for incorporating Twitter into the breakouts and other parts of the event. In the run-up to the show, we’ll blog the whole set.

Thanks, as always, for a thoughtful post and discussion!

I’ve been implementing proprietary interactive meeting tools for over ten years. We’ve basically designed tools and processes that allow for a well-managed “back channel” that’s part of the process for brainstorming, raising critical issues, etc., well before Twitter came along. What I’m consistently surprised by with the advent of Twitter and the backchannels that are created by it, is the lack of design and open dialogue BEFORE the session takes place. So much of what Danah experienced could have been avoided had she said, “Hello everyone. Now I know there’s a backchannel going on and what I’d like you to use it for is to ask me questions, raise issues, tell me what you think. I want to hear from you about the content. It’s hard to talk up here as it is, so I won’t be responding to lighting or how fast I talk. But I will be paying attention to the questions you have.” Then, your backchannel moderator can theme the common questions and issues and make responding in real time doable. Because that’s the other piece that seems to be missing. How do you distill all the data that’s coming in? We work very closely with our clients to design that process so it’s not overwhelming and the frivolous things (“Why isn’t there iced tea offered?”) don’t slow down the event…

I agree with you that the blinding lights were a problem (as was the backchannel being displayed on the screen behind her). Those things rattled her and as a result she started talking too fast. But once that had happened, the fact that she was the only one who didn’t get the feedback was unfortunate.

Thanks for your tips on lighting – very interesting.

Olivia

The ‘but’ is this: the real problem was earlier in your story than you’ve suggested. There is no excuse for lights to be set up to blind the speaker. That’s just sloppy technical work by the venue, and should really have been challenged up front. A spot light which blinds a speaker is too bright for a number of reasons other than that – such as making the speaker look washed out and 2D from the audience’s point of view.

Rather than have lights at that level the venue should be required to provide more lights, each at, say, 60% of the problem intensity. In any case lights should always come in at 45 degrees, not front on.

10 years as a theatre lighting designer wasn’t wasted after all!!

Olivia

I’ve done hundreds of lectures, as many others have, and having thought about this and studied my own sense of how things are going with the feedback I get after, I’m convinced the single best measure of how well I am doing, while I’m speaking, is eye contact.

It’s too hard to sense if they love me or hate me, but if I see a room full of eyes, I’m doing ok. I still have their frontal lobes engaged on me. If I have a room full of wandering eyes, which, btw, I’ve noticed often come along for the ride with yawns, I know I’m losing people and need to pause and collect my energy, give the audience something to do to reengage them, or take something else out of my bag of tricks.

The impossibility of eye-contact in a large keynote style lecture with bright houselights is part of what makes that particular form of speaking so hard. It’s one way. It’s solo. You have have less recourse from guessing about how well it’s going.

There is no way to ever be certain what everyone in the room is thinking, unless you’re in a rock-band, or are a stand-up comedian, as in those forms the norm is high levels of physical engagement with what you’re doing.

And yes, at most conferences only a minority of people are on Twitter. But if those people on Twitter are roughly representative of the rest of the audience, then the backchannel feedback they give may also give a clue as to what other people are thinking.

And although we do get physical feedback from an audience, it’s not always an accurate reflection of what they’re thinking. Say somebody has a very blank look on their face. Are they bored? Confused? Or is that just what they normally look like?

In some cases, Twitter and other backchannels do give you an insight into what people are actually thinking.

Thanks for the link to your post – very useful ideas and insights.

Olivia

To your points: I spoke at the same event, on the same stage. I think one challenge no one talks about is the inverse minority effect. If one person on twitter is confused, but the rest of the audience is doing fine, does it make sense to address that one person’s confusion? Twitter makes it very hard to calibrate this sort of feedback. Whereas eye contact, laughter, posture, and a thousand other things can be assessed at once from the crowd by an attentive speaker.

If you watch the video, which doesn’t show you the backchannel, it’s at worst, a reasonably good talk, I think.

It’s worth asking:

1) What percentage of people in the audience are actively using twitter or any backchannel during the session? At most events its very low. And I’d argue it’s not a representative set of most audiences.

2) Is it worth taking up time when speaking to the entire audience to address that minority, when following up individually later is a better use of everyones time? Sometimes it is (e.g. you are, in fact, talking too fast). But often it isn’t.

I do agree with your suggestions for moderating the backchannel. But I really do think the backchannel is rarely that active, rarely as snarky as it has been in a handful of exceptional cases like danahs’, to warrant this being something most speakers will ever need to deal with.

For reference, my post:

The challenge of using twitter at conferences