For the past three years, Jono Bacon has worked as community manager for Ubuntu, one of the largest and most diverse projects in open source software. Consequently, when he recently published his thoughts on building and managing communities, people listen. More unusually, as I found out in a recent interview, when people like me critique his book, he listens, too, with a view to improving the second edition.
"To me, this is the first part of the story," Bacon says. "The first part's going to have holes in it where it doesn't quite fit the needs of most people, but the second edition will be the part that's really interesting."
Published by O'Reilly Books, The Art of Community available as a print and online book, as well as a free download. It is a mixture of anecdotes and practical advice based mainly on Bacon's experience with Ubuntu. Its topic is community -- specifically, distributed communities in which people interact primarily through Internet technologies such as IRC and mailing lists. Its chapters range from creating and growing such communities to how to govern them and what to look for when hiring a community manager.
The book has received enthusiastic endorsements from other FOSS leaders, and 18 five-starred reviews out of 22 on Amazon.
In many ways, the book deserves this praise. Bacon is an experienced journalist, and few people can match his practical expertise on the topic. All the same, just as you might expect, the book can be criticized on several points, in particular for the lack of context in its general approach and its lack of history.
What is unusual -- and, to a reviewer, mildly disconcerting -- is Bacon's apparent willingness to consider potentially negative comments as feedback. In fact, I could almost feel ashamed for finding faults, except that Bacon's responses and general agreement to my comments suggest that the book is not just valuable in itself, but even more so for the conversations it starts.
A singular perspective
The first thing you are like to notice when reading The Art of Community is that it is a highly personal book. That is a strength, in that Bacon's expertise is both obvious and rare. Yet it is also be a weakness, because the book's credibility is dependent largely on the writer alone.
As Bacon himself says, "One of the caveats of the book is definitely that this is my idea of how I think [community] should work. There's a very strong personal bias."
One problem with this single viewpoint is that, no matter how expert the writer is -- and Bacon's suitability for being the writer of this work is never in any doubt -- is that readers tend to judge a book partly by its citations. Even though The Art of Community, is non-academic work, many of its readers are university educated, and may judge its credibility by its sources.
Admittedly, providing those citations would not be easy, because Bacon's work is in many ways a pioneering effort. "One of the things that was really hard about writing it," he says, "was that there's not a huge amount of content out there about it. There's people who have written books about very general communities, but they talk about it from a largely social science perspective, and there wasn't a huge amount of practical content."
All the same, Bacon does mention some management theory and techniques as relevant, so finding relevant citations should not be that hard.
If nothing else, Bacon might cite community leaders from projects other than Ubuntu. However, when communities other than Ubuntu's are mentioned, such as KDE, Google, and the Linux kernel, they are mentioned from an outsider's viewpoint, with little description of how they are organized internally.
The result is that, although Bacon comments to me that "The thing about community is, there's definitely more than one way to skin a cat," little of this variety is described in the book. Yet, if you are interested in the subject at all, hearing a diversity of opinions would only improve the handling of the topic.
This omission is especially important in the discussion of conflict. As a member of Ubuntu, whose members agree to a code of conduct about how they should interact with each other, Bacon views conflict as something to avoid, or to be resolved as soon as possible. "In my mind, having a more hostile community is less than stellar," he tells me.
I tend to agree with him. Yet, at the same time, I cannot help but notice that members of communities like Debian and the Linux kernel sometimes seem to thrive on a level of abuse that would soon mortally offend me. In fact, the level of invective among kernel developers seems one way that Linus Torvalds' position as absolute dictator is made acceptable, because Torvalds himself is not immune to insults. Somehow, the high level of conflict works.
Responding to this point, Bacon raises the idea that the kernel community interacts this way "because that community is primarily populated by incredibly hardcore developers, who in my experience are generally pretty direct people." By contrast, he suggests that "a community like Open Street Map, which has got a lot more end users, is going to be less direct." So, too, is Creative Commons, whose participants include artists and musicians "who are generally of a less fiery nature." He seems to have a point, but without it, the view of community in The Art of Community can seem too one-sided.
Still another problem with the subjectivity of The Art of Community is that the emphasis is on leading communities, with little discussion of participating as an ordinary member.
"The Art of Community is really written for community managers," Bacon admits. "I wanted to write a book that I could show to other community managers. One thing I'd like to see in subsequent editions is The Art of Community becoming a handbook on how you become a great community member."
For instance, Bacon would like to include more information on the unwritten responsibilities of belonging to an online community. "For example, if you take on an action for something, if you're not going to be able to do it because you're busy, there's an unwritten responsibility that you should let the rest of the community know. Then they can reassign it. If you're a leader, it's stepping down if you haven't got time." Such details would go a long way towards presenting a more detailed picture of the communities that Bacon discusses.
The Issues Today
Another improvement would be a move away from generalities in favor of a discussion of current issues in distributed communities. Talking about his present challenges, Bacon mentions the difficulty of planning in a volunteer community, in which people can come and go, and are under no constraints that might encourage them to complete their work.
Under these circumstances, he says, the twice yearly Ubuntu Development Summits (UDS) used to be better at making plans than following through with them. "UDS would turn into these big idea factories, like the Willie Wonka Chocolate Factory in terms of wondrous ideas and plans about what we could do, but we wouldn't tie that down to specific strategies." Now, he says, he has encourage the idea of detailed road maps in the community, and using Gantt charts and other project management techniques to help realize the plans made at the summit.
"I've got a ton of things I want to write about in the second edition on that specific topic," he says.
Bacon talks, too, of the relative merits of online interaction and face to face meetings, an issue that many projects are dealing with today. Mostly, distributed communities interact online, mainly because traveling can be expensive and difficult. However, he also believes that in person meetings are essential for creating a sense of community.
"People get to reignite those social bonds," he says. "They not only work together, but they also get to play together. Spending time in a bar getting a drink, playing cards with someone, going out, or going ice skating, is so important to helping the community to socially knit together.
"The other thing is, you can get through a lot more content when you're talking and into typing. Talking is way more important than typing, because of the way that we communicate using body language. It's tempting to think that we don't really need that with all the wonderful tools we have on line, but we lose a huge advantage when we don't have that."
At the same time, Bacon prefers summits -- planning sessions -- over conferences, where communication tends to be one way and more formal, as well as bar camps and unconferences, where concrete plans tend to be overlooked.
He also raises the problem that meeting in person places tremendous pressure on people, causing some to become aggressive and others to withdraw. "It can bring all this extra pressure on them that they don't have in online communication."
The trouble, of course, with mentioning such current issues is that they may not be so current a few years later, at the end of a book's life cycle. However, even if the particular issues become obsolete, mentioning them would still give readers a stronger sense of the sorts of issues that distributed communities face.
Lack of context
Something else that I would like to see in The Art of Community is a greater historical perspective. In person, Bacon talks about how collaborative communities began with small groups of developers then spread to include non-developers, but, in the book, the idea of communities is treated in isolation, as though it suddenly sprang into being with no antecedents.
This lack of history is especially noticeable in the discussion of leaders, which says little about how their role has evolved. As Bacon points out to me, he was one of the first community managers on the job, yet in The Art of Community, he only intermittently takes advantage of his unique perspective.
By contrast, in conversation Bacon remembers that, when he started, the job was "about cheerleading. It was about getting people excited, and about shouting from the rooftops. It was very much about motivating people. My job still includes a reasonable degree of that, but my job now is about building management and strategy around the community and helping parts of the community work."
Similarly, while Bacon tells dozens of personal stories, he could usefully talk more about how his understanding of the job has evolved. For instance, when first hired, he says, "I definitely went in with the naive notion that I could have a personal oversight on everything. And that's just completely impractical with Ubuntu. I mean, the community's so big. I have come down to the sober realization that I'm just not going to be aware of everything all the time.
"The other thing is, I was naive in the notion that I felt that most of my community had broadly the same perspective."
Now, Bacon adds, much of his work is understanding the differences in people -- "Learning about how people think, how people work together, and how people feel insecure and motivated about different things. How people work and how that can change -- all those sorts of things."
These changes are relevant to other community managers, because they show what is expected of the role, and, perhaps, how the role changes as a community grows from a small to a large one. Since Ubuntu is larger than most communities, such context might have helped the book become more relevant to small projects.
Starting a Conversation
These criticisms may sound like I see nothing of value in The Art of Community. But, in fact, nothing could be further from the truth. If anything, I value Bacon's book partly because of its limitations and omissions.
The Art of Community is by no means the definitive treatment of its subject. But it is the first treatment of its subject, and that makes it valuable as a starting point for discussion. Even the book's faults are useful as conversational hooks to draw people into the discussion. They are starting points for thinking and dialogues. As Bacon himself says, the book is like the 1.0 release of software -- something that is not perfect but something to build on.
Whether a second edition is eventually released is still uncertain -- and will be for at least a year, according to Bacon -- but that does not change the importance of the conversations it is beginning to start. Those conversations alone make The Art of Community, faults and all, one of the more significant books about free software that has been released.