Learning From (and About) the Backchannel

My personal post on the backchannel at the Microsoft-sponsored Social Software Symposium I just attended is already yielding a range of reactions. Not surprisingly, some of those reactions are critical. The idea of a backchannel can be pretty damn scary—but my sense is that the fear comes most often from people who haven’t participated in one, and therefore are likely to both overestimate its negativity and underestimate its value.

I’ve written before about the modes I’ve observed in the backchannel at conferences, but I don’t think I’ve done a good job of talking about the benefits that accrue to participants in these channels as a result of their participation, or the benefits that “leak out” of the channel both during its existence and afterwards.

Let’s start with why the backchannel springs up, particularly at conferences with people who are (theoretically) there because they want to be there, and therefore predisposed to be interested. (That’s a very different context than a classroom or training venue where students are there because they have to be.)

Seb Paquet notes one reason that a backchannel might emerge in such a setting— to allow “otherwise captive audiences [to] rush inside as soon as things feel too stuffy.” But it’s not just when they’re stuffy—it’s also when it’s something the audience members may be interested in, but have heard before—increasingly the case when there’s overlap between gatherings, and somewhat inevitable when you have a bimodal audience like the one at this symposium, and couldn’t take background and shared context for granted.

The backchannel when an audience isn’t engaged in the presentation—whether or not it’s the “fault” of the presenter—does sometimes border on being rude, and critics are absolutely right to call me (and the other participants in these chats) out on this issue. We may well be allowing ourselves a degree of rudeness by cloaking it in technology, and that bears watching and checking. However, the backchannels I’ve participated in, for the most part, have had a minimum of this behavior, and a lot more that adds value. (In fact, as I skimmed the transcripts I saved from the backchannels at this symposium while writing this article, I found remarkably little criticism or negativity, but lots of rich content.)

So, on to the issue of how these emergent location-based backchannels add value to an experience for the participants—and even for the non-participants.

One advantage provided is the ability for people to respond instantly to interesting ideas, while the topic is still fresh in their mind. In the past, I wrote (some of) these things down in my private notes, which meant that they were useful to me and only me (unless I later wrote them up for my blog). It also meant that if I misheard or misunderstood, I was stuck with my misinformation. When the observations are posted in the channel, however, they provide an opportunity for instant feedback and clarification. Take, for example, this brief exchange from #socialcomp during danah boyd’s presentation on Tuesday afternoon. danah had referred to a community of “burners” on Friendster, which was a cultural reference not all of the listeners were familiar with:

Note that there are actually two conversations going on here; one question being posed about danah’s reference, one following up on an earlier question by the same user regarding blog census information. Both resulted in URLs being posted to the channel, and both served to resolve questions on-the-fly. Because a misunderstood reference can make it difficult to grasp the points the speaker is making, these kinds of immediate responses are very valuable. (Yes, I realize that asking out loud accomplishes this, as well. But many people aren’t comfortable doing that out loud, as any classroom teacher can tell you. Additionally, at this event we were specifically asked to hold our questions until the end of each panel.)

In the comments to my post on mamamusings, Scott Golder suggests that audience members should take notes privately, and save this kind of discussion and clarification for after the talk. But there are several problems with that approach. The first is that the discussions after the talk typically take place in very small groups, often among people who already know each other and have a shared frame of references, so there’s less opportunity for new or unexpected voices to chime in. The second is that in most conferences, the talk time is minimal compared to the listen time, so there aren’t a lot of opportunities to hash out all the ideas that might arise during the presentations. And a third I alluded to above, that waiting to get clarification may have a significantly negative impact on your ability to understand and process the talk that you’re hearing.

In addition to providing real-time collaborative information related to the speaker’s topic, the links provided in the channel throughout the day were available later to anyone who chose to archive the discussion. (As an aside, I also attempted to archive most, if not all, of the URLs posted to the channel over in del.icio.us, under a specific heading of socialsoftware/symposium.)

For people not at the event, real-time discussion (ideally coupled with an audio and/or video stream) allows those not physically at the event to participate remotely, posting questions and related information that can then be used by those in the room, often resulting in questions that might not have been asked in the room.

Another benefit that accrues from a real-time chat accompanying the presentation is the ability to cross-reference transcript portions with times. What were people talking about when Steve Johnson made his interesting remarks about echo chambers vs flame wars? What were they talking about when Mimi Ito discussed untethered social networks? At least three people have asked me to send them transcripts from the open chat channel, specifically so that they can see what people were saying during their talks. The time-stamping that many IRC clients offer (I use Conversation, which posts the time every five minutes in the cat stream) makes it easy to do this.

danah boyd and Clay Shirky and I had a bit of a debate during the second day of the conference about whether the backchannel was exacerbating or reducing fragmentation of the people at the event. danah felt that the IRC channel was creating a problematic split between those who knew how and/or had the tools to access the channel, and those who didn’t. Clay and I argued that the overall fragmentation at the event was actually being reduced by the backchannel, rather than increased.

In the physical context of a conference or meeting (especially one that attempts to bring together diverse populations, or one that brings together old friends who seldom have face-to-face contact opportunities), people tend to cluster with those that they know and feel comfortable with. Colleagues and friends sit at the same table, have lunch together, share insights. It can be very difficult for anyone to work up the courage to break into these physical groups, introduce themselves, try to find common context for conversation. But the backchannel doesn’t have a limited number of chairs. Anyone can join—and as the two-day event wore on, more and more people did. It allowed conversations to occur between people who wouldn’t have known to seek each other out otherwise. In my case, there were at least five different people who approached me during breaks based solely on the backchannel conversation—they were interested in what I’d said, or had information on what I’d asked, and wanted to take the conversation deeper; that simply wouldn’t have happened without #socialcomp.

It is also true that at least one secondary backchannel arose at this event, and that it was a more “exclusive” environment. I’d argue, however, that exclusive is too often considered a pejorative term; I would actually describe it as more “intimate.” It was people who knew each other already, who had shared context for remarks, who wanted to strengthen existing friendship bonds.

The nice thing about these secondary backchannels (and it’s entirely possible that there were others that I didn’t know about, with people not in my close social network) is that they (1) can coexist with the larger public backchannel (this is one of the things I love about IM, too…that I can have two simultaneous “private” conversations going on without overlap or intrusion), (2) can be created and dissolved quickly and easily, and (3) aren’t constrained by physical barriers (if you’ve got more than two people in your intimate network, you can’t sit next to all of them at once).

Compare the “private backchannel” to one of the round tables seating up to six participants in the conference room. Was the selection of who was at a given table somewhat exclusionary? Sure. Are some of the conversations at those table private? Yes. Is that a bad thing? I certainly don’t think so. I don’t have the desire to be part of the intimate social network of everyone at an event I attend—and even if I did, that’s not an entitlement that anyone should be able to demand.

There are many topics I find myself wanting to explore in this space. What factors correlate most with fear or resentment of the backchannel? How does participation in a backchannel change the participants’ views of an event or ctivity? The speakers’ views? The organizers of events? To what extent does the backchannel subvert authority and/or undermine credibility? What happens when we empower an audience to do ridiculously-easy group-forming on the fly in a specific location? Closer to home for me, how will these backchannels change the way we teach? Or the way we conduct routine meetings? (Many faculty in my department, for example, want to ban laptops from faculty meetings because of the growing use of IM between faculty.) And on the tools side, how can we build tools that integrate real-time activity like chats and im’s with notetaking and recording?

This backchannel thing isn’t going away, folks. We have to learn to live with it, to live in it, to thread it into our tools and activities. Or at the very least, we have to learn to tolerate it, and understand that it can’t simply be banished or ignored.