How Nonprofits Can Grow Their Own Collaborative Communities

Larry Sanger

Keynote delivered at the Kintera User Conference, San Diego, California, Feb. 7, 2007.  Kintera builds software for nonprofits.

Donors will give you money only if they trust you, and a standard way to build that trust is by building relationships with donors. And one way to build those relationships is to build online communities. In other words, get people together on the Internet. The hope is that potential donors will develop greater awareness and trust of the nonprofit that is organizing the community--and then be more likely to support the cause.

The good news is that building small online communities is not hard, if you have one or two motivated and adequately knowledgeable people. What's harder, and rarer, is building really big communities.

I was asked here to tell you the story of how Wikipedia got started, and how the new Citizendium project is getting started as well. So I'll do that, and then I'll draw some more general lessons about how to set up collaborative communities. I will conclude by offering you some rough guidelines that your nonprofit might find useful in spearheading new collaborative communities.

1. How Wikipedia got started.

I don't have time to tell you the whole story about how Wikipedia got started. So I'll focus on the parts that concern project organization, or in other words, what we did to get it started.

One thing many people don't realize is that Wikipedia emerged out of a pre-existing parent project, a peer-reviewed free encyclopedia project called Nupedia. So we already had a mailing list with a couple of thousand e-mail addresses. Having that list helped a lot in kicking of Wikipedia. We had a ready-made group of people, many of whom I already knew, who were already interested in making an encyclopedia. The situation with your nonprofit might well be the same: you've got the ear of a lot of people you know who are interested in your cause. So you might be able to mobilize them online the way we mobilized the old Nupedia contributors. That's one reason that I'm actually quite optimistic about nonprofit organizations starting up, say, information-gathering wikis. Many of you have already got a motivated, informed, mature community. That puts you way ahead of most twentysomething computer programmers.

Anyway, it took a heck of a lot more than our local village. We did more recruitment. We posted to a lot of online mailing lists--I mean, e-mail discussion forums--and tried to get new recruits that way. Online communities are very frequently built by the online equivalent of word of mouth: personal e-mails, mentions on forums, blog posts, and the like. Obviously, if you can interest the media in your project, that helps, too. But, actually, we built the Wikipedia community without any significant media mention; the first news report about us came nine months after launch. Anyway, the point is that you must actively recruit participants for your project. It is not the case, with collaborative communities, that if you just set up a wiki or a discussion forum or similar "social software," people will just start using it. You have to get the word out there.

Anyway, we started with a fairly small group of people who were already interested in making an encyclopedia. We expanded that group partly through active recruitment. But we had to overcome another problem faced by very many projects: how do you get the people who have merely declared their interest actually involved, putting in large amounts of time on the project? It's one thing to gather people together; it's quite another to get them to do something. That's the problem of motivation. Well, the solution to the problem has at least four parts: openness, which attracts the eyes of new people; ease of contribution, without which people just won't want to take the time to get involved; free information, which gives them an altruistic reason to get to work; and strong collaboration, which both provides people with much-desired personal visibility and amplifies their efforts. Let me talk about how Wikipedia illustrates these concepts, and why they're so important.

The first concept is openness. This is simple, but crucial. It means that anybody can come to the wiki and look at any page. You don't have to pay a fee to read it; you don't even have to sign in. The reason this is important to community-building is that the single best way to attract new contributors is to show them the products of your old contributors. New contributors need to see that something is happening, something that they want to be part of.

Second concept: ease of contribution. This is one thing that makes Wikipedia famous, or infamous. It's the "encyclopedia that anybody can edit." This is literally true. Anybody can go to most pages on the website and press the "edit" tab, and immediately start editing an article, without even signing in. And editing an article is, technically, about as easy as writing an e‑mail. "Wikiwiki" means "fast" in Hawaiian, and it was called that for a reason: it really is fast and easy to change the contents of a wiki page. The ease and rapidity of contribution is essential to community-building, because every seemingly small hurdle to involvement that you put up will cause large numbers of people online to just give up and never come back to your website. People expect things to be easy, or they won't get involved. Just think about your own case: do you spend a lot of time trying to figure out complicated systems so that you can join an online community? Probably not.

Third concept: free information. Here I don't just mean information that is free to read--I mean information that is free for anyone to reappropriate, to reprint or develop further. In other words, this kind of freedom is something like being in the public domain. (I don't mean the same as the public domain, but I won't bore you with fine distinctions right now.) The point, anyway, is that the information is not proprietary--it is not exclusively owned in any traditional sense. The reason this is important is that, if you ask your volunteers to create content that is owned exclusively by some company or organization, a lot of them will refuse. They don't want to work to line anyone's pockets in particular. If they freely contribute their own information, then why should anyone in particularown the information? The point, then, is that you'll have more participation if people know that the whole world, and not just the organizers, can use and reuse the information that people contribute.

The fourth and final concept is strong collaboration. Wikipedia isn't just collaborative, it's strongly collaborative. When I say that, I mean that not only are there multiple authors, and not only are those people each others' editors, but beyond those things, there is no set group of people who are the authors and editors of an article. Some authors drop out, and other authors join. In five years, you might not know any of the people who are maintaining an article that you started. The collaboration on Wikipedia is particularly radical because many of the collaborators are anonymous. I don't recommend that as a policy, but it is a common feature of collaboration online. Now why is strong collaboration important? For a couple of reasons that I can think of. First, when other people take the time to edit your work, it gives you a real boost, it increases what we might call "personal visibility." It's an honor. It also sets up a mild sort of competition. I say "mild," because even if you're trying to one-up each other, you still share the goal of making a better article. Collaboration also is what makes Wikipedia as surprisingly robust as it is. You might think that something so open and easy to contribute to would be full of vandalism. But the more simpleminded vandals are actually shut down very quickly, in most cases anyway, because so many more people are doing positive work on the project, and they care very much about it.

So I'm saying that these four concepts explain how a large group of people was not only gathered together, but also motivated, to work on Wikipedia. The website was open to all; it was easy to contribute to; people loved the idea that it was free, so that the world would always benefit from their work; and they collaborated, egging each other on.

This might explain why people were motivated to contribute to Wikipedia. But, you might wonder, why did it take off exponentially, as it did? If you look at a graph of the number of articles Wikipedia has, you'll see that it's been pretty much exponential growth. It has grown so fast that it now has over 1.5 million articles in the English language alone.

The reason for that is something called the Google Effect. At least, this is the explanation I've always believed. As you know, Google's computers are constantly scanning the Web for new sites and updates to old sites--this is called "spidering." So here's what happened. The first time Google spidered the Wikipedia website, we got a big spike in traffic, because bunches of people would search for information on some specific topic, and Google would send them to the Wikipedia article. More traffic overall meant more people at work on articles. So we created not just some new articles, but more new articles than we created before. Then Google would spider the website again, and traffic spiked upwards again. It was what we might call a "virtuous cycle." Or in other words, the website grew virally. New articles were like a virus that went forth and "infected" (or brought in) new people, who "incubated" (that is, wrote) more of the "virus" (articles).

Obviously, not all collaborative communities can grow virally the way that Wikipedia did. Wikipedia had the great advantage of creating something that people were searching for: general information. This you can take as another lesson from the Wikipedia experience, namely, if you want to attract a community, have the community create stuff that people actually want to look at.

2. How the Citizendium is getting started.

Let's shift gears now, and illustrate a few more principles from a project I'm working on right now--called the Citizendium, or the Citizens' Compendium. Let me first explain briefly how I got from Wikipedia to this new project. Perhaps this will speak to some concerns you might have about collaborative communities--in particular, concerns about quality and reliability.

When Wikipedia got started, it was in a tandem with its more responsible parent project, Nupedia. But after Wikipedia got started, Nupedia was allowed to wither and die, leaving the younger upstart without any approval process and without any official expert editors. Now, of course, people often used to criticized Wikipedia, from the very beginning, saying that it would produce garbage that people couldn't trust. My reply always had two parts. I would say: "First, Wikipedia has a remarkably robust system. If you actually look at the articles, you'll see that they aren't that bad at all. The reason for this is that far more people want to make a positive contribution to the project than a negative one. They then correct each others' work, and it is easy for us to delete or roll back changes that vandals and other troublemakers add." But then I would say, "Second, the end result will be an article that can quickly go through Nupedia's article approval system. Then we'll have a large set of articles that are unquestionably reliable. Nupedia and Wikipedia are an unstoppable high-quality article creation juggernaut!"

It was the second part of my answer that tended to shut most people up. The problem was that, as I said, Nupedia withered and died. Why did it, you ask? Well, the company that funded these projects got hit very hard by the Internet collapse of 2001-2, and so they had to lay off all nonessential personnel. Since Nupedia and Wikipedia were making no money for the company, they had to lay me off. Wikipedia was able to survive without me, but Nupedia was not.

Essentially, then, Wikipedia went it alone. I was Wikipedia's chief organizer in its first year, but the project was never fully in my control. Many of the new recruits indulged in some immature and disrespectful behavior. The project developed a culture of anarchy, as you might have expected it would. It really became a reflection of a broader Internet culture, which I think is really an extension of youth culture. Other aspects of this broader Internet culture are anonymity and a rejection of the notion of expertise. So it was a serious faux pas for you to support a point on Wikipedia by saying, "Well, I have a Ph.D. in this field, and this point is totally basic to the field, so I know what I'm talking about." You might be conversing with a high school student, who will say: "I don't care what your credentials are. Prove it to me." The Wikipedia community will support the student and say your credentials are irrelevant.

Now, for most of you, and in general for responsible organizations run by professionals, it goes without saying that experts deserve special roles, and that collegial behavior should be required, in any community you'll have anything to do with. This may sound obvious to the people in this room, perhaps, but it isn't very obvious to many of Wikipedia's contributors.

For a long time I thought that Wikipedia would eventually reform itself. Events of the last few years, however, have persuaded me that Wikipedia will never make the sort of fundamental changes it needs to make, in order to become a truly mature community and a reliable source of information. This by the way is another lesson we've learned about online communities: once they're started, they tend to get quite set in their ways, because they are self-selecting. There's a relatively brief window of opportunity in which you can change course in some important way; after that, if you want to do things very differently, you have start a new community.

So, last September I did start a new community, and a new project, that I hope will correct Wikipedia's problems. It is called the Citizendium, which stands for "the Citizens' Compendium." The Citizendium community is, like Wikipedia, aimed at creating a giant free encyclopedia using a wiki. But we also have a special role for expert editors; they can approve articles and gently direct the development of articles. They don't have to approve every edit; rather, they work together with rank-and-file authors just as people do on Wikipedia. The difference is that editors' expertise is respected and, when necessary, they can settle content disputes. Another big difference is that we require the use of real names. We also take a no-nonsense approach toward community disruption: if you don't respect the basic rules of the community, you aren't welcome in the community. In other words, we want to foster a culture of real-world responsibility.

The root motive behind the Citizendium project is to give the world an expert-guided, reliable, but still enormous free encyclopedia. We also want to create a community that is more inviting to be a part of.

So--you might wonder--how are we doing? Well, we've been at work on the Citizendium pilot project for about three months. It has been a private pilot, only because we don't have enough Internet servers to handle the Internet traffic we expect. We hope we will soon come into the donations and in-kind support we need to launch publicly.

Despite the fact that we're not open yet, we have over 160 expert editors and over 350 authors--all volunteers--and we just yesterday added our 1,000th active article. So it seems to me that it's going quite well indeed, and it's only going to get better once we open up and start enjoying the Google effect.

Let me review now. I said there are some principles that are critical to the success of collaborative communities. I've mentioned the importance both of starting with a good group for launch, and also of doing recruitment after that. I've also said that Wikipedia took off because of at least four features: openness, ease of contribution, free information, and strong collaboration. But Wikipedia, like many other collaborative communities online, suffers from various problems. And these problems, in my opinion, are avoidable. For instance, you don't have to permit anonymous contribution in your project. Also, you can make special roles for experts, even in projects that are otherwise quite egalitarian and "flat" in management structure. So, basically, I would have you develop collaborative communities that are both interesting and attractive to contributors, but also which produce a lot of reliable information for users. Why aim at anything less?

3. Ideas for some projects nonprofits could start.

Next, consider a few ideas about some types of collaborative communities you might start. I'll begin with the simplest and easiest sort of community, and then talk about something more complex and ambitious. After all, your nonprofit doesn't have to aim at anything as ambitious as Wikipedia in order to build a "sticky" community that generates interest and loyalty.

The simplest sort of community would be either an e-mail-based discussion list, or a Web-based forum. In other words, a way for people to get together and talk to each other. A discussion group isn't really collaborative, but it can be a sort of community, and it can create a lot of interest--so it might be a good first "baby step" toward organizing your contributors in a more robust way. In my opinion, discussion groups are one of the best uses of the Internet. They can be enormously informative and interesting, and they can be the basis for long-term personal relationships.

It's also easy for any reasonably competent system administrator to set up mailing list or Web forum software for you, and the software is usually pretty low-maintenance. But, obviously, setting up the software isn't enough to make a successful discussion group. To create the critical mass needed for an ongoing, lively discussion, you'll want to spread the word about your group, and make sure you have one or two people who are quite active and ready to say some provocative things. Another important point is this. You should pick the topic and name of the group carefully--a topic with wide appeal, but which does not simply replicate the topic of some other active group. You need a niche.

Now if you want to organize a collaborative community, one of the best ways to do that is to set up a wiki. Again, it is neither difficult nor expensive for a competent system administrator to set up a wiki, but wikis are much harder to manage and make work. There are zillions of "dead wikis" out there that no one is working on anymore. So, maybe the most important choice you'll have to make about a wiki, if you want to set one up, is the choice of what its purpose is. Notice I don't say merely the topic of a wiki, but its purpose. While discussion groups are set up for the purpose of talking about some topic, wikis can be set up for many other purposes. That's because wikis generally involve gathering information, but you can gather information for many purposes. For example, do you want to educate the public? Document software? Record best practices or how-to guides for the use of professionals? Create a specialized reference database?

In fact, it's meaningless in a way to say, simply, "Let's set up a wiki," if you haven't got an idea of what the wiki is for. It's kind of like saying you want to hold a conference without a clue about what the conference will be for. You should set up a wiki only if there is something in particular that you want to do.

Well, what can you use a wiki to do? Assuming that you want to create a robustly collaborative community for your nonprofit, I have three questions that you can use to come to an answer yourself.

Here's the first question: "What kinds of information are of interest to everyone in the field that your nonprofit serves?" The greater the interest in the information, the broader the interest in your wiki. So, if your nonprofit does cancer research, for instance, it might be good to have a wiki that documents cancer research; but it might be better to have a wiki aimed at the end-user about important advances in cancer treatments, as that is no doubt of broader interest.

Another relevant question is this: "What sorts of information do people collect or review, which other people collect and review in pretty much the same way?" Things like bibliographies, summaries of important articles, and general introductions to a topic are written and compiled over and over again by different people for their own personal (or organizational) use. It can be helpful to have these people work together, then, rather than endlessly duplicating effort. Again, if your nonprofit does cancer research, then your primary constituents probably review the news of latest developments in treatment in pretty much the same way. Since that's something that many people are already tracking, then, again, you can bring people together by having them work together on summaries of those latest developments.

My final and perhaps the most important question is: "For each of those types of information, what sort of information resource can be created only using methods of broad-based, wiki-style collaboration?" In other words, I'm asking: what sort of information resources can people in your field create only if very many of them work together as parts of a huge, ever-changing, rotating group of information maintainers? After all, if a magazine, or an individual, is already perfectly capable of doing a good job doing whatever you propose to do with the wiki, then people probably won't be very well motivated to work on your wiki. It would be embarrassing, in a way, because you'd be competing with professionals who will always be able to do a better job. Let me explain how I would think this through. To change the example, suppose that you are raising money for education in a third world country. Then I would ask you: is there some information resource regarding third-world education that could be created only using a strong collaboration? Perhaps the information you want to amass is just too vast, and not profitable for any for-profit company to compile.

Well, I can think of an example. You could keep an ongoing log of all available news articles, from all sources you can find online, about third-world education. Then your collaborators could submit summaries of those stories. It would not be just a blog, which summarizes some of the news--anyone could do that. It would be a massive undertaking, to summarize every significant news story about third-world education, far beyond any one person or just a few people. But if you get enough people who are passionately interested in the topic together, they might egg each other on and perhaps even compete to see who can create as many summaries as possible. And by the way, for this particular project, you wouldn't necessarily have to use a wiki, although using a wiki would certainly pose many advantages. You could also make it a blog. Blogs too can be used for collaboration, if many people share posting rights to the same blog.

I could easily multiply examples, but I'll wrap this up. Actually, one of the great things about collaborative communities, and all this Web 2.0 business generally, is the sheer quantity of ideas. Online communities that are more or less collaborative have actually been going on virtually since the beginning of the Internet. We've seen a rather brutal Darwinian sort of competition among the projects. Most of them die out because of fundamental flaws with their design. But others have survived and will no doubt go on surviving for a long time. Give us another ten or twenty years, and this vast experience will allow for a virtual science explaining how to create a successful collaborative community.

In the meantime, though, there's going to be a lot of trial and error. So, if you're the adventurous type, I'd encourage you and your nonprofit to give it a shot. You can help your organization, you can help your constituents, and most importantly, you can help your cause.

By the way, if you do decide to try to start a collaborative community, I'd be very interested to hear about it. You can find my coordinates on my personal website, larrysanger.org.

Thanks very much.

Home