The challenge of visible twitter at conferences

Last week I gave a keynote at Web 2.0 expo NYC, and as you can see in the photo below, one of the interesting things this year was the twitter feed for the conference was placed on stage behind the keynote speakers. Any tweet with #w2e was put up live, on stage, a few seconds later. The slides, if any, speakers used were placed on the large screens to the left and right of the stage. I’ve seen twitter on stage, and presented with it up before, but never right smack dab behind me on stage.

I’ve spoken at Web 2.0 expo before, and as in the past, the event is well run. As a speaker I’m treated well, given a walkthrough of the stage, a microphone and slide run through.. a lot of stuff is done to help me do well up there and these guys spend money on things only speakers see. They hire a producer to manage the stage, provide a big confidence monitor up front so I can easily see my slides while presenting, which helps tons. It’s a high quality event as far as I’m concerned as a frequent speaker.

But the cognitive science here is simple: anything in motion on stage takes attention away from anything else. Even if I were JFK or MLK, if you put a moving screen behind me, with blocks of twitterness appearing randomly every 20 or 50 seconds, it can’t possibly help anyone understand anything. It divides attention away from whatever it is the speaker is trying to do. Unlike pop-up videos on VH1, this isn’t an 80s video you’ve seen 100 times that needs spicing up, at any time, by anyone with a twitter account, in the room or perhaps even not.

An hour lecture is one thing, but for short 10 or 15 minute keynote sessions in front of an audience of 2000+ people, which is what these were, the cognitive science is working against doing anything that the speaker isn’t orchestrating him or herself.

For danah boyd, who spoke a day earlier, the mirror in mirror effects of allowing the audience to have control over the stage worked against everyone involved, including the audience . It hurt danah’s ability to do what people came to hear her do, and gave a minority of the audience a minor round of momentary, but ultimately forgettable snarkishness instead. (Read her thoughtful post about what happened here).

Now I’m all for snarks. In fact I’m all for people speaking up when I’m not doing my job as a speaker. It is a kind of feedback, and if I can hear the complaint there’s a chance I can do something with it. But If I can’t even hear you, who exactly are you complaining to? There’s a 0% chance I’ll get the message if I can’t see or hear it when it’s delivered.

Speaking in public is hard enough. And while I’m all for people having their say, and getting involved, putting words up on a screen behind me is not an effective way to do it. At a minimum, put the twitter feed up somewhere I can see it, on a side screen, so there’s a 10% chance I’ll figure out why everyone is laughing when I don’t think they should be. At least a speaker might have a fighting chance. Or enlist the host to watch, and let them tell me if I’m going to fast, or too slow, or doing something easy to fix.

To Brady and Jen’s credit (Web 2.0 co-chairs), they did make an announcement the next day and take responsibility for what happened, switching the feed to moderated. Speaking up at the event itself was brave and cool of them to do. They, in essence, took the back-channel chatter about the back-channel issues, and brought it up to the frontchanel. Postmoderm indeed.

If you’re thinking of using visible twitter feeds as part of your event consider this:

What problem are you trying to solve? If you want more interactivity, you must either make the sessions longer, smaller, or have a Phil Donahue type person who can field questions from twitter, or the audience, in real time and pass them on to the speaker. If your audience gets bored, get better speakers. Or train them. Or perhaps buy them all a good book on the subject from an author who writes about this stuff).
Help speakers, don’t work against them. At nearly every event, the speakers are the agenda. We are your primary draw. Even if we suck, people come because you listed us as the reason to come. In the green room backstage at Web 2.0, I asked nearly every other keynote speaker if they thought the stream behind our heads was a good idea. Not a single one said yes (of course they’d all seen danah’s experience, so we benefited from her rough ride). If we’re the agenda, work with us, not against us. If you want us to speak better, or be more interactive, then ask us or train us in the arts of public speaking.
Break sessions in half and interact some dialog in the middle. Matt Mullenweg cleverly divided his 60 minute session at web 2.0 in half, taking a break for questions in the middle. Why not do the same for 20 minute keynotes? at the ten minute mark, have the host pull one question from twitter, email, or smoke-signals, and ask the speaker. If everyone knows this is part of the routine, there will be less pressure on the back channel – they know there is a path for their comments or questions to hit the front channel. This can be tight for 10 minute keynotes, but it’s preferable to giving back-channel control the entire time.
Some people like the back-channel staying in the back-channel. If I’m joking around, as if I was passing notes in high school, I don’t want my snide comments broadcast. I’m sure some folks at #w2e made a new tag, #web2realbackchannel to get around this, or more likely, simply didn’t use twitter at all, since they didn’t want their words, and twitter handle broadcast to the room.
Interaction doesn’t have to be real time. The real thing people want is a chance to interact with the keynote folks. So why not have a designated time when they’re all around in the same place, say at lunch? Or have an email address anyone with a question for a keynote speaker can send, anonymously, and post the answers to the event blog? All of these solve the core problem, of being interactive and participatory, without any of the drawbacks of competing for attention simultaneously with the speaker.

Despite what happened, I’m a big fan of experimenting with public speaking, and finding ways to make it a better experience for everyone involved. Brady & Jen get some points for trying something new, and making adjustments when it caused problems. But changes can be driven by what we know about cognitive science, attention, and be a collaboration between speakers and organizers. I’m very hopeful dialog around all this will spin in that direction, instead of simply giving up and banning things without examining what can go wrong and why, or what the real problems are we’re trying to solve by having speakers up there in the first place.

(You can watch my web 2.0 expo keynote here)

Posted Tuesday, November 24th, 2009 at 3:26 pm

in public speaking | 18 Comments » |

Dave Gray - November 24, 2009 at 4:59 pm

Hi Scott,

This is an excellent, thoughtful post. I agree with you fully that the backchannel should be unmoderated and it shouldn’t be set up in such a way that it detracts from the speaker.

I do think that there might be speakers who want to take a more conversational approach who might want, and benefit from, a visual tweet stream that lets them interact more dynamically with the audience.

I also have seen panel discussions where a live, visual tweetstream improved the dynamic. I think this is partly because in panel discussions, there’s always one person on the panel with enough bandwidth to monitor the stream — sometimes a moderator — and this can create an interactive dynamic that would otherwise be impossible.

As a speaker I find the backchannel to be a great help at improving my speaking ability, but I make sure to wait till after I am done speaking before I read the tweets. Tweets like “When will this guy get to the F-ing point?” and “Now it’s getting better — THIS is interesting!” are very helpful to me — good, solid, actionable feedback.

We’re still figuring out Twitter and many other social dynamics that are impacted by technology. There are bound some well-intentioned efforts that go wrong like this one did. But that means we’re learning. I wouldn’t have it any other way.

Scott Berkun - November 24, 2009 at 5:12 pm

Dave: I’m hoping to find some kind of summary of different uses of twitter at conferences, and the feedback/report from the organizers after the fact.

So far I’ve received lots of anecdotal emails from folks at various events, but it’s still highly speculative.

Cliff Atkinson has a new book, and maybe he dug up some stronger stuff:
http://www.amazon.com/Backchannel-Audiences-Twitter-Changing-Presentations/dp/0321659511

Olivia Mitchell - November 24, 2009 at 5:50 pm

Hi Scott

There is some research on the way people use Twitter at conferences. Here are two papers that I’m aware of:

http://www.scribd.com/doc/16287533/More-than-just-passing-notes-in-class-The-Twitterenabled-backchannel?autodown=pdf

http://lamp.tu-graz.ac.at/~i203/ebner/publication/09_edumedia.pdf

They’re both have a fairly small number of respondents – but it’s a start.

And Maury Giles has done some analysis of the content of live tweets:

http://consumercentric.biz/wordpress/?p=18 and http://consumercentric.biz/wordpress/?p=63

Hope that’s helpful.

Olivia

Steve - November 24, 2009 at 5:55 pm

Scott! Great post. This topic has been bouncing around a lot lately. Thanks for pushing it.

I spent a chunk of today defending my critical (albeit supportive in certain dynamics) position on backchannels on Twitter and my blog. Basically, what Dave Gray said regarding certain dynamics working for it is where I stand. Many presentation dynamics are horrible, or sketchy for it, at best.

Most of the time a backchannel is up against brain research on cognitive processing, and amplify that combat if you have a giant screen directly behind you while attempting to deliver a complex and engaging message.

Scott Berkun - November 24, 2009 at 6:23 pm

Awesome – thanks much Olivia. I’ll check these out.

Still looking for the definitive tip sheet for organizers in how to use twitter to everyone’s advantage. Cliff’s book might be it, but we’ll have to wait a few weeks to see.

Scott Berkun - November 24, 2009 at 8:24 pm

Joe: fair enough. I think there are ways to make it work, but the speakers have to be in control and comfortable whatever it is.

Moderation in some way seems the way to go.

Olivia Mitchell - November 25, 2009 at 1:28 am

I have written an eBook to help presenters with the challenge of the backchannel during their presentations. You can access it on my site.

Joe – A live twitterfeed displayed on the screen can work when it’s under the presenter’s control and it’s an integral part of the presentation. For example, the presenter asks a question and asks people to respond via twitter (or other backchannel tool). Then the presenter displays the responses on the screen and discusses them.

There are two tools: http://www.backnoise.com and http://www.todaysmeet.com that people can use immediately without having to set up an account. That means people who are not on Twitter can also contribute via the backchannel (they do of course have to have a smart phone or laptop).

What Not to Display - November 25, 2009 at 9:03 am

[...] Not to Display Posted on November 25th, 2009 by admin in Backchannel stories Scott Berkun writes an excellent blog post here about why it’s simply bad idea for hosts (or anyone) to display a live Twitterstream behind a [...]

Mike Nitabach - November 25, 2009 at 9:09 pm

That is the most outlandish thing I have ever heard vis a vis giving a presentation. I would flat out refuse to present if there was going to be some kind of uncontrollable editorial ticker streaming on a screen behind me as I spoke. Dana Boyd’s decription of her experience was gutwrenching. This kind of audience-driven and embraced public shaming and humiliation is disgusting. I fail to see why any conference organizer would want to create a situation like this, and why any speaker would tolerate it.

BTW, wasn’t there some kind of similar streaming live-commenting imbroglio a few years ago involving one of the Movable Type founders?

stephen lead - November 26, 2009 at 2:02 am

At our Ignite events we use an unmoderated Twitter stream – but only in the breaks.

I would never consider putting this up while someone was speaking, let alone directly behind them…..

Kudos to the organisers for changing this once they realised it wasn’t working

Scott McArthur - December 1, 2009 at 5:45 am

Totally agree that we should continue to experiment. However, on this I cant think of any good reason for having Twitter running live during a keynote. This goes against good presentation technique. The audience should be listening and watching the speaker who in tun is supported (where necessary) by visuals, props etc. So unless your keynote was about Twitter I think using it was probably a mistake.

On the otherhand opening up Q&A sessions to live Twitter/Skype etc is a great idea. We tried this at a HR 2.0 event in Romania this year and it worked a treat – we had questions from all over the world which really added to the event experience for speakers and the audience.

Public Speakign Course - December 3, 2009 at 8:41 pm

Thanks for sharing your thoughts on speaking with Twitter.

I have never thought it would be much fun for the speaker, and you have confirmed it.

Generally, what is the value of the Twitter board? If you haev 20 people per minute commenting while you are speaking, what is the point of you speaking?

Twitter on stage will only be here a short time before speakers start saying that they want it off or they wont speak.

Twitter is great, just leave it for the appropriate times in a conference. Have sessions where the conversation can continue through Twitter, but not while a speaker is speaking.

Cheers

Darren Fleming
Australia’s Corporate Speech Coach

Alex Bowyer - January 5, 2010 at 11:16 am

I definitely agree we need to explore the use of these technologies to see what works and what doesn’t. But overall I think displaying a Twitter stream behind a speaker takes too much control and focus away from the audience.

In the interests of continuing the dialogue on the subject here is an article I wrote exploring the same issue and the pros and cons around using Twitter during a live event:

http://www.bitcurrent.com/why-twitter-should-never-take-center-stage/

danah boyd : Ce qu’implique de vivre dans un monde de flux | traffic-internet.net - January 6, 2010 at 4:12 am

[...] Rendre visible le fil Twitter des conversations dans une conférence publique n’est pas si sim… qui a écrit un livre sur la prise de parole en public. La projection du flux Twitter peut-être bénéfique dans certains contextes, mais nécessite de connaître l’objectif qu’on lui demande de remplir et que cet objectif soit partagé par tous. Jeremiah Owyang avait rédigé quelques conseils sur le sujet suite à la conférence SXSW 2008 où Mark Zuckerberg, le président de Facebook, avait été confronté au même type de problème (voir également le livre d’Olivia Mitchell Page sur la question). Pour Joe McCarthy cependant, la relation demeure inégale entre le conférencier et son audience. Une inégalité qui ne doit pas donner au public le droit d’être impoli ou de descendre en flamme l’orateur. Mais le lynchage en ligne par la foule sur Twitter (baptisé Tweckle par Marc Parry) est difficile à contre-attaquer. Si la pratique du mur de Tweet se généralise, on voit qu’il nécessite quelque règles : une charte de bonne conduite et la présence d’un modérateur actif pour tenter de contrebalancer les effets de foules, ainsi que la possibilité de pouvoir débrancher le mur de Tweet si nécessaire. [...]