Visionaries

But what is a wiki, anyway? Perhaps Erick Schonfeld of Business 2.0 explains it best when he writes in his column, What the Heck Is a Wiki?,

A wiki is a group blog that can be edited by its readers (as opposed to a regular blog, which one person writes and everyone else reads). Wikis are more like conversations. The few well-known wikis include Wikipedia (the online encyclopedia where anyone can submit an article or edit an existing one – read Wikipedia: Localization in a Free Content Community), Wiki Travel (an amazingly comprehensive travel guide written by travelers), and Wikibooks (textbooks written collaboratively). But most wikis are limited to private groups and are used to manage corporate projects or to collaborate in a quick and dirty way with colleagues.

A wiki is usually started by someone in charge of a project who is looking for collaborative feedback on his or her idea. Startups like Socialtext and JotSpot provide hosted wikis for as little as $9 a month (free in some cases). Instead of drafts of contracts, marketing materials, or lists of potential hires being distributed as e-mail attachments, the information can more simply be posted to a wiki. There, other people on the team, who are all invited to the wiki and have passwords to access it, can read it, add to it, and make corrections if necessary. As chaotic as this may sound, there are built-in safeguards: Every version is saved, and every change can be tracked back to who made it. The ability to link to other webpages is what makes the wiki itself a social document (interacting with other documents).

Ross Mayfield, LISA Member and CEO of Socialtext, goes on to explain, “What we are doing is blurring the distinction between reading and writing. The beauty of a wiki is that it starts like a webpage, then people discover that they can edit it like a Word document.” It gets interesting when people realize they can add Web links to the wiki.”

Mayfield was recently interviewed by Tony Perkins and Rich Seidner on the AlwaysOn blogging network. We offer you a portion of the interview in an effort to encourage you to investigate the workflow and worldwide collaboration possibilities of this social networking technology.

AlwaysOn: How does a wiki work?

Mayfield: With a wiki, unlike the one-way web, anyone can walk up to a web site and make it two-way. You click on or enter "edit this page" and start typing away. You don't have to know any HTML or any code. You click "Save," and you're back from edit mode to display mode.

AlwaysOn: Where did that name come from?

Mayfield: Ward Cunningham, this really cool guy, who used to be on our board of advisors, invented a hypercard-like thing for the web for this community of his around Portland. He was sitting in the Honolulu Airport trying to figure out what to call the thing. He saw the Wiki Wiki bus go by and said, "Oh, that's as good a name as any." In Hawaiian, "wiki wiki" means "very quick." So, he originally called it WikiWikiWeb—it's been shortened by people to the term "wiki." So, that's where it came from.

The cool thing is that it was ten years ago. This is an old open source technology that's gradually saturated the online programming community to get at its sheer utility—nothing else. Nobody's pumping it. Nobody's promoting it. We've had more than a million downloads of source fields alone—of open source wikis. The major platforms are going to 200% a year. You can go to any major IT department or division in any large company, and there will be a wiki, where people are communicating about projects and writing documentation. So, just like blogs, it started a nice cultural revolution that's been going on about how to use this really dead-simple tool and its rise in popularity. And these things have been entering the enterprise from the bottom up.

AlwaysOn: Why did it take so long for people to kind of embrace it?

Mayfield: Well, simplicity is kind of a funny thing. If something in the enterprise world is really, really simple, people don't take it seriously. I mean, this is the classic innovator's dilemma, right? Big companies don't take notice of things that seem too simple and too cheap to be really disruptive to their brands or whatever their products might be, right?

Basically, what's happened is this has been a tool—a tool initially for programmer-like geeks and hackers. What we saw was this great opportunity where, on the one hand, you have this great demand pattern that's taking off from the bottom up inside all these enterprises. So, we asked, "What if you made a wiki even simpler?"

And that's the real disruptive idea. What if we made it easy enough to use not just for people with programming skills, but regular people, anybody? And we made it something that could fit into the existing enterprise infrastructure; the substrate of all these other systems. So, the boring stuff we have to do—like integration with the enterprise directory, monitoring, backup, storage, and search solutions—makes it something that becomes as great a utility as e-mail is, but only for communication between groups.

AlwaysOn: There's been a pattern of people increasingly uploading stuff and posting to the web. Is that just a natural evolutionary way for them to get to know the web? An "Oh gee, I can go on and find a bunch of stuff that's been posted, and now they've made it easy, so I can post and share too."

Mayfield: That's exactly what's going on. If you think about it, we just spent the last twenty years or so developing this physical infrastructure. Now, the challenge is building the social infrastructure. We've had this critical mass of users online, and now we have a decent amount of broadband. Out of that, you get some very complex behavior.

Think of it this way. The geeks and hackers were doing this and that; they used the infrastructure as a social tool for communicating on boards, e-mailing, using mailing lists, and so on. The interesting part isn't just that they were communicating and collaborating, but that given enough time, they were liable to invent something. In this case, they invented open source, a whole new model for producing things that wasn't driven by market structures with price signals or firms with contracts. Instead, you had this self-organizing glob of people—well, not entirely self-organizing—driven by social incentives. It was an entirely different thing that motivated, helped organize and produced goods—whether they were information goods or otherwise.

And so, if you look at what's happening, first as software and open source are disrupted, you have the same darn thing. It's what you're doing in media, right? What do we have? We used to have this monolithic process by which you could get any kind of news—a broadcast network, right? In some way, that's moving to more of a menu-to-menu network, right? Suddenly, all that's happened is that the costs of personal publishing, forming groups, and a couple other things that let more complex social interaction happen—all these costs are falling toward zero? So, the media world's experiencing the same thing, whether it's politics or the Dean campaign or other voteable industries; the same similar disruptions are happening.

AlwaysOn: Can the media business survive, as we've written about exhaustively? The way I look at it is that every publisher is competing for mindshare.

Mayfield: Yes.

AlwaysOn: And it isn't necessarily an apples-to-apples game. Sometimes, you are competing with the movie on the airplane. But clearly, for lack of a better description, with a newer, open kind of media, alternative sources of content are gaining nurture. And we all know there are lots of reasons for that. I think the best, most bottom-line way to describe it would be that people are voting for a more authentic type of content; content that's more specific to their interests and more genuine. So what you're saying is that in the corporate marketplace, there's already an RC business that relies on cooperation. What happens to behavior? How does it change? Where is it trending from a corporate collaboration setting?

Mayfield: When we started off more than two and a half years ago, we thought a lot about the utility, and its premise was, "How do we help small groups of people communicate and collaborate effectively?" Now, we're at the point where we have lots of cases of customers accelerating their project cycles by a quarter, just by shifting their communication from a more hub-and-spoke model to getting it all on one editable page.

Now, some of the really interesting stuff's happening. We're up to about a hundred customers. Twenty of them are Fortune 500s. We have people ripping out their intranet and replacing it with a dynamic wiki. What does that do? It's a 30-day process to get anything posted through the webmaster and other gatekeepers, technical or editorial. So, the people are just setting the dynamic wiki utility up themselves. Then, really interesting stuff starts to happen.

Before, everybody had the same portals for three, four years. They haven't been updated, since they were purchased and some initial energy went into them, right? All the value of that information is gone. You end up with one big tree-and-branch structure emanating from a pole. There is no link structure, for example, that a group of clients could suck up and take advantage of.

The difference is that when you reduce the barrier to contribution, and you entrust the users to work with their own information, you end up increasing the amount of information that's available. It increases the probability that, if I'm searching for something, it might actually exist. The other advantage is that you're getting greater diversity. And that's the beautiful thing. More diverse domain expertise is being applied to different places.

You see it in the Wikipedia. We have collaboration on a scale that has never happened before. This is part of what's new. You have millions of strangers trusting one another, sharing and creating a common resource. That same thing is starting to happen inside companies now. What that means is, instead of domain expertise about who is the Archbishop of Canterbury or whatnot, it's about a specific business process that's otherwise hidden. This is the kind of tacit knowledge that otherwise has only been tracked in email. Suddenly, it's being shared in publishing as media. Most of it is just people having conversations more openly. The same communication activity that occurs in email can, as a byproduct by doing it through a wiki, develop a really rich knowledge base that anybody can tap at anytime.

AlwaysOn: I hear that people say this really cuts down on email because you're not emailing everybody separately.

Mayfield: I'd put it this way. You have a problem in email called "occupational spam." It's when you carbon copy or blind carbon copy; you stretch your email into a broadcast medium. Google says that's about 30% of email. We found the same measure.

The interesting thing, though, is that when you give people an alternative—a way in which they can communicate from one person to a whole group, or even in menu-to-menu fashion, and a space designed for menu-to-menu interaction—you get some simple efficiencies. Email is a point-to-point network, with this big mess. A lot of people are playing email volleyball. Nobody knows which attachment is the latest version. You're not communicating very clearly, efficiently, or effectively.

The difference with a wiki work space is that it's private for a group. I send an email into the work space; it gets turned into a wiki page. I can tag it. I can route it to a blog. Then, when I'm consuming information, what I can do.... Actually I'm not even consuming; I hate that word. I can then choose what I want to subscribe to by using an RSS feed. Do I subscribe to this project log because I'm a member of the project team? Or do I not? And I have the choice to, or not, because there is still more transparency. So when I have downtime, I can go through and explore the periphery of all my information. That's a better way.

AlwaysOn: If you subscribe to a wiki RSS, are you getting notified every time somebody posts something?

Mayfield: You have a choice. You can choose: "do I want to subscribe to recent changes?" which is a feed of every single edit, every single change. Or, as a subset of that space, there are RSS feeds being generated for each tag and for each blog.

For example, I may be active on some projects and need to be notified on everything, and there may be other projects where I'm not. I may have multiple wiki workspaces and have control, not just through RSS, but through email notifications. In one workspace, you might want to be updated the second something happens; in another, maybe once a day; in another, maybe once a week. And that's a pole model of attention management.

Think about email. It's all just information being pushed into your inbox. You've no control; no unsubscribe button. So, what the pole model does is let individuals choose how to manage their own information flow, which is something that system designers can't do more optimally than the individuals can themselves.

Editor’s Note: You can find the original blog postings and the rest of the interview at Why Wiki?, Kill Your E-mail, Wiki Wiki!, Big Brother Is Watching, Open-Source Kwiki Equals Money in the Bank, and Will Wikis Enter the Mainstream?