Despite declarations of "Web 2.0", the "Read/Write Web" and other trendy nomenclatures, there is very little that is new about wikis or any of the so-called social media so prominently positioned in magazines and web sites. The rules for a successful wiki are very similar to the rules that one should apply to any community site (that's what we used to call them because we started to call them social media sites). One could argue that a wiki is a modified forum (it retains many forum features).

The following section contains advice based in large part on my years of experience managing community-oriented sites where the users are the main contributors of content, much like wikis. I also owe a large debt to Christian Wagner and Ann Majchrzak, whose paper Enabling Customer-Centricity Using Wikis and the Wiki Way provided me with some particularly useful insights (Christian Wagner and Ann Majchrzak. Journal of Management Information Systems/Winter 2006-2007, Vol 23, No. 3, pp 17-43)

Their research was focused on enhancing constructive customer engagement in a wiki, which is not necessarily how most companies will use one. However, the principles they suggest are good ones that, in my experience, do in fact foster a sense of community and collaboration. In their review, they compared wikis operated by the Los Angelse Times (thinly veiled with the name "Boomtown Times"), Novell and, of course, Wikipedia. The Los Angeles Times wiki was a dismal failure, the Novell wiki, a moderate success, and the Wikipedia wiki was, of course, a smashing success.

Wagner and Majchrzak offer up six propositions based on their research, but I have taken the liberty of condensing them into four rules of thumb, based upon my own experience with managing collaborative web sites.

Alignment of Goals

Wikis got their start being used by programmers to document software projects. This is an ideal use for a wiki because there is a strong incentive on the part of the programmer to participate. There is a social contract at work. The software needs to be documented and all participants must be kept in the loop so one programmer keeps his documentation updated with the understanding that other programmers will do the same in a mutual back-scratching arrangement.

This apparent no-brainer is, apparently, not a no-brainer. At least it is not a no-brainer to the editors of the Los Angeles Times, who decided that it would be good to have a "

" - an editorial composed and edited by the masses. The LA Times started by posting their editorial, and then provided a wiki for the public to respond. As Wikipedia has learned, there is no alignment of goals among political types and the partisans reigned supreme during the very brief life of the wikitorial. In the organic, evolving world of wikis, consider the wikitorial an evolutionary dead-end. At first, the users made a good faith effort to collaborate on an editorial, but they soon concluded that producing a single editorial that was acceptable to everyone was not going to happen and there had already been attempts to delete the entire editorial, so by the second day, they forked the editorial so that it would be possible to represent different points of view. Once news of the wikitorial experiment showed up on Slashdot, a technology-related news website (http://slashdot.org), it attracted a lot of attention and it was soon followed by pornographic posts and so on. On the third day, the wikitorial was shut down.

The kind of vandalism encountered by the Los Angeles Times represents the nightmare scenario that almost always gets raised as an objection to using a wiki. In fact, the first time I saw a wiki I thought it was a lousy idea for this very reason. As it turns out, this kind of defacement is not as common as one might think, and when a wiki is set up and managed properly, can largely be avoided.

In the late nineties, I was responsible for what we called “community publishing” sites. It wasn’t a wiki per se, but it had many of the features of a wiki, the most important one being that we (the newspaper) used the Internet to let the community participate in the publishing process. The site was called NCHometeam.com, and it was a partnership between WRAL-TV and The News & Observer in Raleigh, North Carolina. It was a statewide high school sports site. Coaches and interested parents were recruited statewide to update the rosters each season, and then to update scores after the games on Friday night (which often finished too late to get in the paper and there was not enough space).

We quickly learned that the coaches were enthusiastic participants, and they were just as committed as we were to making sure the information was timely and accurate.

In retrospect, it’s easy to see that fears about false sports scores being posted were unfounded. While coaches did have an incentive to win, they had no incentive to cheat and post false scores because there were so many other people at the game that they would easily get caught. The transparency of the process meant that it was in everybody’s interest to post factual data. In other words, the goals of the entire community were aligned.

The reason the Los Angeles Times wikitorial failed is that an editorial is a point of view about a controversial subject. The goals of the individuals on either side of the debate are to discredit those who disagree with them and establish their world view as pre-eminent. In other words, the goals of the left and the right are not aligned. Therefore, I suggest no bi-partisan wikis. Ever. There is no such thing.

This does not mean that everyone who participates in a wiki has the same goals, nor do they involve themselves in the same activities. By goal alignment we mean that their goals are not in conflict; they all head in basically the same direction.

A Culture of Collaboration

In practice, the most common problem encountered by new wikis is that it can be difficult to get people to participate. There are quite a few psychological and organizational barriers that need to be overcome. Most important, in addition to needing to have the technical apparatus to operate a wiki, you also have to have an organization with a culture that fosters collaboration. If you don’t, it is likely your wiki will fail to thrive.

For example, a certain government agency has decided to launch a wiki that will capture all of the undocumented but highly useful information that floats around in people's heads. They are facing two sources of internal resistance.

First, they have a hierarchical culture where every communication gets approved by proper channels. Being propositioned by some young twenty-something about brain-dumping your wisdom into a wiki after having every utterance scrutinized by your superiors for your entire career is like suddenly being told by your wife of twenty years that she thinks you should loosen up a little and get a girlfriend.

The second source of resistance is the fact (or perception) that once your brain gets dumped, it becomes communal property and while you may be fairly certain that you will continue to get your paycheck, you are not so certain that you will continue to get credit for your faithful fidelity and the cultured wisdom you have nurtured for so many years. Owning information is a source of power; that's why it's so hard to get people to share it.

In one case, a department was more than willing to post content in a content management system, but they were unwilling to do the same work if it was with a wiki. The reason? They feared a lack of control. It can represent a loss of ownership for people. When people are rewarded for individual output, they are going to be less inclined to participate in a project with collective output. While they may not say this out loud, what they are worried about is whether they will still get credit for their good ideas and hard work.

Universities are also experimenting with wikis. Wikipedia is such a wild success, you as the professor reason, that it would be fantastic if you had a wiki for your class. Then instead of letting students write papers for an audience of one (which is you) they can write them for their peers - their fellow students - and for posterity - all the students that will follow. Therefore, a classroom becomes a source or repository of knowledge. You even dreamily fantasize about students correcting and expanding upon the postings of other students in a communal editing effort where everyone is both the student and the professor. The only problem is that students have no interest in correcting (or updating or expanding) another student’s work. What's in it for them? They annoy a potential date and don't really have much to show for their work. There are two reasons to go to college - to learn and to acquire documentation that you have learned in the form of a transcript or diploma. I've used papers I've written in grad school as part of my portfolio. What kind of portfolio do I have if it was all done in a wiki?

Most of your school life is spent being told to do your own work and keep your eyes on your own paper, while being forced to read wordy honor codes and the like. Traditionally, schools have not fostered a collaborative environment, so students aren’t quite sure how wikis fit in with the culture of the school. Wikis are an excellent tool for the classroom, but you will need to be prepared to help the students unlearn some of what they have learned about what is appropriate behavior in school, just like employees of the government agency will have to relearn what’s appropriate for them as well.

The one common theme that runs through all of these examples is that the goals of the participants were not aligned. In some cases, their respective goals were in direct opposition to each other, while in other cases, there was a belief that participating in the wiki would not provide enough individual benefit. If you want to derive a benefit from collaboration, you need to make sure that everybody in the organization also benefits from collaboration.

Community Custodianship

I have already mentioned that when I first worked on community publishing sites for a newspaper, we conscientiously avoided creating the appearance that we were the “editors” of the content. In this case, we were doing so in order to avoid liability for what was posted on the community sites. This meant that we didn’t actively monitor user posts and that we wouldn’t remove posts unless a member of the community raised a concern with us.

What we had done inadvertently was to shift the monitoring responsibility to the community itself. Again, this was not for any altruistic reason, such as a belief in decentralized decision-making. As it turns out, letting community sites be managed, in effect, by the community is an important component of successful sites.

Despite the open nature of wikis, an effective wiki is not an egalitarian free-for-all. Just as the members of the community share reading and authoring privileges, they must also share custodianship of the community. The community rules the community. In this custodial role, the community of users needs to establish rules of conduct for contributors to the site, and they also need to monitor user activity, to make sure that it is in conformance. The custodial role means that users are not only responsible for identifying suspect content, but they also serve on the decision-making bodies that establish guidelines for when such content gets deleted, or when users get banned.

Clearly Defined Rules for Posting Content

Successful custodianship means that in order to get your users to participate fully, your wiki needs to have clearly defined rules and processes. These rules include a clear description of the kind of content that should be contributed to the wiki as well as rules for how to handle disputes. For example, Wikipedia has “five pillars” that define the character of Wikipedia, and here is a sampling of a few of the rules:

The rules are very explicit and leave little room for ambiguity. These rules and others can be found at the following locations:

In addition to establishing rules, you need to seed your wiki with content when it is first launched. The presence of content will facilitate the creation of even more content. One of the advantages of seeding the wiki prior to opening it up to a larger group is that the pages that you create serve as a kind of template for the new users to refer to when creating their own pages. In other words, they serve as an example of the kind of content you want to see on the site and having seen an example, people will be more comfortable producing their own content for the site.

Monitoring User Behavior

When Ronald Reagan talked about nuclear arms reduction treaties with the former Soviet Union, he espoused the following philosophy: “Trust, but verify.” Running a wiki requires trust on the part of management in the capacity of their employees, their customers and the community at large to behave reasonably well, most of the time. Since it is not realistic to believe that they will behave reasonably well all of the time, then you must switch to “verification” mode and monitor behavior. Despite the definition that wikis are sites that anybody can edit, the truth of the matter is that if you let just anybody edit it and do not, at the same time, provide a mechanism for proper oversight, your wiki will not work.

The ability to monitor user behavior creates transparency and transparency is good. The very fact that behavior can be monitored will keep most of the behavior that needs to be monitored from ever happening. In fact, while the most common objection managers have to using wikis is fear of vandalism, the biggest problem they end up having is just the opposite: no activity at all. The monitoring requirement varies with how widely available the wiki is. In other words, a workgroup wiki behind the corporate firewall needs less monitoring than would a customer-accessible wiki that the public can see.

It is also important that the monitoring and policing of the wiki remain the responsibility of the community. As I said earlier, community custodianship is one of those factors that creates well-run wikis and one of the roles the community plays while acting in the capacity of custodians is the role of monitor. Not only should the community itself be the monitor, it should also be the body that helps to determine what the rules are in the first place.

Monitoring behavior can be more than simply a policing role. One of the reasons employees can be reluctant to participate is a fear of losing credit. If anybody can edit a document, how am I going to get credit for writing this one? Since most wikis can now track changes (MediaWiki can), you can also monitor activity on the wiki as a means of identifying good uses of wikis. This is especially true in educational settings where students might be graded on their activity.