This is the first time ever that I cared about SXSW conference or was jealous for not being there. Watching the blogs and Twitter stream, it appears to have been better and more exciting than ever. I guess I'll have to figure out a way to finally get myself there next year....
But this post is not really about SXSW. It is about presenting at such conferences. More specifically, how the back-channel (on Twitter and elsewhere) affects the way one needs to approach an invitation to speak at meetings where much of the audience is highly wired online: to say Yes or No to the invitation in the first place, and if Yes how to prepare and how to conduct oneself during the presentation.
After the meeting ended, Jay Rosen described in great detail all the things they did to prepare for the session and how that all worked - go and read: How the Backchannel Has Changed the Game for Conference Panelists. I will be sending the link to that post to all the speakers/panelists/presenters/moderators at ScienceOnline2011 once the program is set. That is definitely a post to bookmark and save if you are organizing a conference, or if you are ever invited to speak at one.
This includes people who tend to speak at conferences that are not filled to the brim with the Twitterati. Even at such conferences, a small but loud proportion of your audience WILL tweet. Be prepared! Even if you are speaking at the AAAS meeting.
There are other important things to think about - both for organizers and presenters.
First, public speaking is for some people the most terrifying thing they can ever be asked to do. But even those who are not completely terrified, may need some training in order to do well. Have new people be mentored by experienced speakers (I mentioned how we do that at ScienceOnline at the end of this post) by sharing the panel. As an organizer, work hard to help the new speakers to alleviate their fears, to make crystal-clear what is expected of them, to provide them support before, during and after their sessions.
Many organizers are hoping to increase diversity (of personal experiences and approaches, not just in terms of gender, race, age, ethnicity and such, though the diversity in the latter usually brings along the diversity in the former as well). They need to remember that announcing this intent is not enough. People who were not welcome at the table before have no reason to believe that they will be welcome now - so why bother. You have to do more - actively reach out and engage them. And, as your conference (like ScienceOnline) goes through years, if you are successful at bringing in the diverse groups to the table - they will notice. They will invite others to come next year. The meeting gains reputation, over the years, for being open and inclusive (nobody is a superstar and everybody is a superstar). Instead of being tokens, they become an integral part of the conference and help shape it. This takes work.
Second, the Back-channel should never become the Front-channel!!!! Never display tweets on the screen behind the speaker. Never. On the other hand, please make it easy for the speakers to monitor the Twitterverse on their own computer screens if they want to.
Third, if you are organizing a conference, think hard about the format. At a typical scientific conference, the speaker is a scientist who is presenting new data. The talk is likely to have a level of complexity (as well as an arrow of the narrative) that is not served well by constant interruption. In such cases, a traditional format, with a Q&A period (long enough!) at the end is just fine. TED and TEDx conferences are similar. Quick presentations, like Ignite or storytelling events are similar - the presentations are too short and too well-rehearsed to be able to withstand interruptions. But you have to have a Q&A at the end - it is irresponsible not to have it.
For example, many sessions at the AAAS meeting are three hours long! Including my session. And in each one of those that I attended, the moderator announced at the beginning that the Q&A will be at the end. Hmmm, how many people will still be in the room after 2 hours and 40 minutes? They will be either long gone, or brain-dead and eager to leave. So we tried to do the best we could with the format we had - we had 2-3 people ask questions after each one of our presentations (there were six of us at the panel) as well as at the very end. And you know what - at the end of the third hour, the room was still full and we were still getting more questions. Engaging the audience early on got them excited. They wanted to stay in the room and engage some more, 2-3 of them every 20 minutes or so, and several more at the end.
On the other end of the spectrum to the one-to-many lecture is a fully-fledged Unconference format. It is based on the insight that "The sum of the expertise of the people in the audience is greater than the sum of expertise of the people on stage.". This, of course, depends on the topic, the speaker, and the audience.
As I explained at length in this post after ScienceOnline'09, and at even more length in this radio interview after ScienceOnline2010, our inaugural meeting in 2007 was a pure Unconference, but that we since decided to move to a hybrid format for a number of reasons I explained in both of these places.
Think, for example, of Workshops. We had a Blogging101 workshop at a different day, time and place in 07 and 08. We expanded the number of workshops in 2009 and had them as a part of the main program (just tagged as workshops), and then in 2010 we again moved them all to a different day, time and space to make it clear that these sessions are different - not Unconference-y in format, and for a good reason (we'll do the same next year).
A Blogging101 workshop, for example, will have an experience blogger at the front. The audience will be full of people who have never blogged and want to learn how. The moderator is an expert, and acts as a teacher or trainer or 'fount of wisdom' to the audience who came to get exactly that - instruction. The audience expects to learn how to start a blog, how to post the first introductory post, how to make a link and insert a picture, how to build a blogroll and change the visual design of the blog from an existing set of choices. They also expect some sage advice on what is regarded as proper blogging behavior so they do not get instantly slammed when they enter the blogosphere for "doing it wrong". The kinds of questions such an audience asks are going to be calls for help and clarification, perhaps for more information. They are unlikely to insert their own opinions and information, or to challenge the session leader. It is more of a classroom lecture (or lab) than a freewheeling discussion. Yet is has its own usefulness and should not be looked down upon because it is not in an Unconference format.
Actually, a Blogging102 workshop, where the audience already has some experience in blogging and is looking for tips and tricks for making their blogs better, looking better, and promoted better, there will be additional insights from the audience - which we saw at Scio10: that workshop was quite participatory and interactive.
Then, there are demos. A demo is just 12 minutes long with additional 3 minutes reserved for Q&A. The presenter is showing off his/her website or software or what-not to people who have not seen it before and would like to see how it works. Again, interruption of such a short and carefully prepared presentation would not be a good thing. If you have more to discuss - grab the presenter in the hallway afterwards. We are thinking of moving the Demos (both 12-minute presentations and potentially stations or booths) to a different day/time/space next year as well. Nothing wrong with that format, but it is not in an Unconference spirit.
Yet, the bulk of our conference is an Unconference. And we have seen that well-prepared presenters can turn even large 4-5 person panels into lively discussions off the bat. I have described one such 2009 panel in this post and there were several this year (most notably the Rebooting Science Journalism session). What we tell both moderators and participants is that the name of the session is not a title of a lecture but the topic of the conversation for that hour.
People who already have experience with the unconference format lead the way (we try to have such people lead the first morning sessions to set the tone for the rest of the event) and n00bs follow. Once everyone is in the swing of things and participating freely, it is easy to have a session be very informal. For example, last year Pete Binfield and Henry Gee started off their session with the question "Our topic is "A" - what do you want to talk about?". And that worked brilliantly as people who decided to attend that particular session already had questions and comments prepared in their minds and were ready to start discussing the topic right from the start. Other sessions require more of an intro, and that is OK as well.
So, the bottom line is that there is a spectrum of potential formats and each format has its pros and cons. The duration (from 5 minutes to 3 hours and everything in-between) will dictate how participatory the session can be. The relative difference between the expertise of the people on stage and the people in the audience is also a factor - more even they are, more participatory the session should be. As an organizer, always strive to have the sessions as participatory as the format/topic/people allows it, not less. Having less will diminish the experience - it will be seen as preaching down and trust will be lost.
And keep the Back-channel in mind - people in the room are not the only people participating. Make sure that the people following on Twitter, or Ustream or SecondLife can participate to some extent as well - perhaps let the people/audience in the room (all of them or a few chosen individuals) be moderators of Twitter chatter, and ask the cameraman to introduce questions from the Ustream audience into the room. We did both at ScienceOnline2010 and the feedback from virtual audience was positive. We'll try to do even better next year.