The first day of the summit attracted an astonishing 200 people, including leaders throughout the computer field (especially its free software side) as well as interesting representatives of other communities. The preponderance of free software proponents can be explained by its link to the Open Source convention (because one of the sponsors, O'Reilly Media, obtained space for the summit in the same hall) and the background of organizer Jono Bacon as community manager for the Ubuntu project.
A word about the organization of an "unconference" is in order for those who have not experienced the pleasures of such an event. An unconference, like a conference, brings a range of people together in one location to explore issues and learn the current wisdom in a particular field. But instead of granting conference organizers the power to choose speakers, it allows each participant to propose a topic.
The logistics of an unconference are impressively simple and deceptively powerful. Attendees are greeted by a large whiteboard in a central location with a list of conference rooms and times--all blank! It is up to the attendees to fill the slots. Although some organization can be done in advance on the web, the board is empty until attendees come on the site and fill in the slots.
The Community Leadership Summit was billed to "bring together community leaders and managers and the projects and organizations that are interested in growing and empowering a strong community." Jono Bacon created it as a follow-on to his new book The Art of Community: Building the New Age of Participation (which I co-edited, and wrote about in a recent blog).
Many attendees were old hands in the free software movement, while others were just people trying to understand communities which they found themselves heading, often outside the software field. Topics fell into two broad categories, as I saw them:
- Focused topics on achieving immediate goals. These included motivating community members to contribute, implementing strategic plans for communities, and so on.
- Broader topics in the area of communities. These included recognizing ethic diversity, fostering open source in developing nations, and encouraging more women to enter computing fields (which includes participating in computing communities).
The full list of sessions and summaries can (eventually, after participants add their notes) be found on the conference wiki.
The two sessions I proposed fell in the latter category. One concerned the importance of organizing citizens into communities to participate in the new efforts of governments to be more open, a theme I covered earlier in several blogs (for instance, one proposing adaptive legislation). This session drew only six participants, but I was satisfied. Each had engaged in serious work on some community project or political activity and was very serious about making government work for its citizens.
My other session was on educating community members, an area I have researched and written on extensively. This session drew eight participants, and provides a good example of the agility one needs to run an unconference session. I was all prepared to talk about documentation and training for technical communities, but most of the participants were in an entirely different space. Their problems were more basic and insistent, usually rotating around getting new members interested in sticking around.
The eight of us were extremely diverse, including a project that attracted new members all the time but lost a lot of them right away and another project that had lost half its members over the past few months. To spice it up, we had a community manager whose online community was devoted to preserving cultural artifacts of an indigenous people in Mexico, numbering about 25,000 and concentrated in two villages, with a language so unusual that some of its characters haven't been assigned codes in international standards.
We had a lot of fun, we lost only one attendee (who spent time with me later going over his own concerns), and I think I successfully conveyed my research as well as pulling insights from the rest of the attendees.
An unconference can be a range of things. Some sessions can be simply another way to arrange a conference. Nothing prevents an expert from coming with a prepared slide show and doing a conventional, frontal presentation. At the other extreme, someone can propose an important topic on which no one at the conference is an expert, including himself, and can lead a fascinating exchange of ideas. In other words, an unconference makes the most of the special talents and experience of the people who attend.
Kaliya Hamlin, an expert in leading unconferences, gave me a bit of their history. The idea arose some 20 years ago but was mostly constricted to narrowly defined meetings, often held by a company for its own staff.
O'Reilly organized one of the first modern unconferences, featuring a wide-ranging selection of attendees and a relatively open agenda. This event was the first FOO ("Friends of O'Reilly") camp in 2003. I was privileged to attend this unconference, and blogged about it.
FOO camps have taken place annually since than and expanded, branching out for instance into Science FOO. Because attendance at FOO camps is carefully controlled to make sure everybody is a leader of some sort in his or her field, they predictably led to a reaction called BAR camps, and then a general interest in unconferences.
The unconference model worked spectacularly well for O'Reilly, with our extensive contacts in technology. Clearly, any attendee who chose to talk about the work he or she did every day would have something compelling to offer, and the hallway discussions that are the highlight of every conference would be particularly productive here. But could the FOO concept work for open conferences where attendees are self-selecting?
Experience shows that it does. Certainly, this weekend's Community Leadership Summit was a tremendous success. I sometimes found myself in wonderfully pregnant situations--for instance, the moment today where I was sprawling out on the floor like an eighteen-year-old college freshman with one of the leaders of the database field who had expressed an interest in documentation, pulling up articles of mine for him on my laptop.
Few people enjoy this experience, needless to say, at a standard conference.
The Community Leadership Summit was nominally free and open to all. Anyone in the Silicon Valley could just drive in (and many took advantage of this convenience), whereas others flew in from other continents to participate. (Of course, most of these people had other reasons to be here--usually the Open Source convention.) But we recognized that social and economic barriers still prevented many from coming. Downtown San Jose is not a cheap place to stay or eat in, and living or travel here is reserved for those who can afford it.
Consequently, Jono and the organizers of this summit will try to organize regional community leadership summits around the country. Jono also urged attendees to add their notes to the conference wiki so people who couldn't make it could still get something from our talks.
Certainly, an unconference is just as good as the people who attend. People need the courage (or desperation) to suggest session topics that matter to them. They need to respect others while expressing their views, and moderators need the standard facilitation skills: keeping the conversation focused, moving it along, squelching efforts to dominate discussion, encouraging participation by everyone in the room, and synthesizing statements from different people.
But an unconference, in my experience, seems very hard to "game." If someone suggests a topic that reflects too much of his self-interest, he'll end up in an empty room. If he pursues his personal agenda without respect for others, he could start out with a full room and end with an empty one.
Locating the Community Leadership Summit right before the Open Source convention led to inevitable comparisons, which I found unnecessary. I believe that unconferences will expand in popularity and frequency. On the other hand, there will still be a role for the Open Source convention, which I believe to be an exemplar of the best of the traditional conference model: stimulating keynotes by people with unusual things to say, useful tutorials that bring attendees up to the level to benefit from the sessions that follow, plenty of chances for attendees to chat intensely with speakers, etc. And did I mention the parties thrown by sponsoring companies with free alcohol?
I'll end with a comparison made by Jeff Osier-Mixon, a community manager in charge of the embedded Linux community Meld. He's new to community management as well as unconferences, and joked that he'd like to say he's running a user community, but actually is walking a user community. Having recently reread the classic book by Eric Raymond, The Cathedral & the Bazaar, Jeff saw an unmistakable comparison between the metaphors in Raymond's title and the contrast between a conference and an unconference.
Our world is increasingly a Bazaar process, and unconferences are likely to grow in their appeal. As more people attend them, the attendees will learn the techniques that make such gatherings work and improve them even more.