It’s a Wiki Wiki World

My third column is up at AlwaysOn. It's about the world of Wikis with commentary on the recent joust between Ross Mayfield's Socialtext and Joe Kraus's JotSpot and the issues surrounding Wikipedia. Check it out!

It's a Wiki, Wiki World
Skirmishes and issues abound in the Wild West atmosphere of the emerging social software space

The first month of 2005 had both the tech world and the blogosphere buzzing about wikis. For those unfamiliar with the term, a wiki is a website (or other hypertext document collection) that allows users to not only add content but also edit any existing content -- a definition I pulled from one of the best wikis around: Wikipedia, a free, online collaborative encyclopedia.

It all started a few weeks ago with the (to my mind) overblown "wiki war" that erupted between social software companies Socialtext and JotSpot after the latter lured Disney away from Socialtext to become its first major customer. Founded in 2002 by (now CEO) Ross Mayfield, Socialtext -- which received $500,000 in angel funding from leaders in the social computing space -- provides enterprise social software to more than 50 organizations. JotSpot, in contrast, is the newcomer to the sector: Co-founded by Graham Spencer and Joe Kraus (who also co-founded Excite), the company recently closed a Series A round of $5.2 million from Mayfield Fund and Redpoint Ventures, and is about to come out of its beta launch.

Things can get heated in the blogosphere, and true to form, the primary players weighed in with their own (sometimes cranky) comments about the so-called war. Socialtext's Mayfield wrote the following in his blog:

We're glad to be the market leader that JotSpot feels it needs to goafter ... Zero-sum thinking is lazy and, quite frankly, old school. It cements itself into the business model, product and culture of a company very easily ... Jot may think it has won a skirmish against its leading competitor. But really, it's at the cost of its customer ... The only harm to us is taking my time to write this post."

In response, JotSpot's Spencer posted the following in his own blog:

"One thing in Ross's post merited a comment here: He says that we're 'old school' for thinking that we can convince developers to build on our closed-source platform. I'd like to suggest instead that it's 'old school' to focus exclusively on the openness of the code when the openness of the data is at least as important. Aren't Flickr, Technorati, Google, Amazon, and eBay closed-source platforms, and haven't they spurred some of the most interesting 'innovation at the edge' that we've seen in the past couple of years?"

While the war of words is amusing, it also seems beside the point since the companies don't truly operate in the same market space: Socialtext is leading and creating the enterprise wiki space (which allows for more efficient collaboration and communication among corporate teams), while JotSpot is using a wiki-based program to create simple software applications (many of which seem to be spreadsheets with increased functionality for tracking competition, managing deals, and call-center support).

I sent the JotSpot link to a friend who's an IT consultant at IBM, and he loved it. He was looking for a simple customer relationship management (CRM) program to assist a nonprofit in managing membership and outreach, and JotSpot provided the perfect solution.

All of which reminded me of a meeting with Marten Mickos, CEO of MySQL, in which he talked about the "law of nonconsumption," explaining how MySQL brought databases to people who'd never dreamed of using them. The result: MySQL was downloaded more than 12 million times last year. JotSpot could pull off a similar feat by making CRM programs (as well as other apps whose high costs have traditionally limited them to corporate customers) accessible to the masses. Although some people -- like Om Malik, a senior writer at Business 2.0 magazine -- believe that wikis will never represent more than "micro-niche" products, JotSpot's nonconsumption market may just surprise a lot of people.

Wikipedia: The quest for openness and accuracy. Another debate raging in the mainstream press and blogosphere revolves around the validity and reliability of the very site I mentioned at the beginning of this column -- Wikipedia. Robert McHenry, former editor-in-chief of the Encyclopedia Britannica, led the charge, stating that Wikipedia's contribution and editing processes are flawed and providing examples of entries that have been edited into mediocrity.

The Wikipedia FAQ states the following:

Some unspecified quasi-Darwinian process will assure that those writings and editings [sic] by contributors of greatest expertise will survive; articles will eventually reach a steady state that corresponds to the highest degree of accuracy.

To which McHenry counters, "Does someone actually believe this?

Although I use Wikipedia, I agree -- at least to a degree -- with McHenry's criticism of the site. Online collaborations like Wikipedia need order, structure, and a voice of authority to weed out the junk and trolls, and to increase the overall quality of the information they contain.

According to Larry Sanger, co-founder of Wikipedia, a core problem with the site is "the dominance of difficult people, trolls, and their enablers." Says Sanger, "I might have continued to participate, were it not for a certain poisonous social or political atmosphere in the project."

What did he expect? Wikipedia is a great, accessible resource that should be considered the first step in serious information gathering or a quick dip in the pool when time is short. When I visited Wikipedia for the first time last year, I found myself laughing at the George W. Bush and Republican Party entries, where you can see the countless back-and-forth edits from both sides of the spectrum.

In response to Sanger, Clay Shirky, an adjunct professor in New York University's Interactive Telecommunications Program, wrote the following on the Many2Many group blog on social networks: "Of course librarians, teachers, and academics don't like the Wikipedia. It works without privilege, which is inimical to the way those professions operate ... [As to] Sanger's final point -- that the Wikipedia is anti-elitist -- yes, it is impossible for experts on a subject to post their views without molestation, but that's how wikis work.

Danah Boyd, a Ph.D. student at University of California, Berkeley, who's researching articulated social networks (and is a co-contributor with Shirky to the Many2Many blog), counters Shirky's comments by stating, "I would argue that many librarians, teachers, and academics fear Wikipedia (not dislike it) because it is not properly understood -- not simply because it challenges their privilege."

Human nature is on display at its best and its worst in forums like those mentioned above. It seems that the smaller the community -- due to interest or barriers placed -- the less trolling and deterioration in the quality of conversation and collaboration occurs. In some of the blogs I visit, however, you can see this deterioration in quality as comment strings lengthen and tempers flare. The result is that often times the best voices turn away because they don't want to deal with the trolls and troublemakers.

In wide-open systems like Wikipedia, there's not much entropy -- but there's not a high level of order either. What does exist is something akin to a Lord of the Flies world order: Battles and skirmishes occur among the children, and the naval rescue squad flies in to save the day. Much of the criticism of Wikipedia has focused on whether the site's Arbitration Committee and processes have, in fact, served as an effective rescue squad.

While the term open source has been used loosely to refer to a number of software development trends, there's a big difference between open-source software development and open-source collaboration. Open-source software has succeeded in part because of its innate order and structure -- rules and standards that developers must adhere to so that their code will be readable by everyone. Wikis, in contrast, are based on words and perspectives, not code, and there are no barriers to entry.

Recently, I compared my experiences living in Korea to those of a friend who's been living in Russia, and found that we'd both drawn similar conclusions -- that people living in environments with few laws and little enforcement tend to act in a more lawless manner. In both societies, the resulting disorder and inefficiency could be seen in everything from basic traffic violations to overtly corrupt business practices.

The lesson? It's a wicked, wicked world out there, and everyone needs rules to live by. Wikipedia is no exception: Marked more by inconsistency than entropy, Wikipedia (like other wikis) needs more rules and enforcement if it's to improve as a source of information.

In the Many2Many blog, Shirky states, "Wikipedia makes no claim to expertise or authority other than use-value. If you want to vote against it, don't use it. Everyone else will make the same choice for themselves, and the aggregate decisions of the population will determine the outcome of the project."

I disagree. In a wide-open world without any review standards, how does anyone identify what's valid and true? You can find a professor within just about any "accredited" university to substantiate almost any view -- no matter how radical, fringe, or unsubstantiated. Which leads to the question, How do you provide checks and balances in the far less structured environment of the web? Some of the responsibility must fall on those that provide the information -- especially if that provider positions itself as an encyclopedia.

My concerns about content validity, however, extend beyond Wikipedia to the problem of accessing information in a world overloaded with data. Greater access necessitates greater discernment. As public repositories of information such as Wikipedia proliferate, the ability to filter that content will become a necessity.

Ideally, this responsibility should fall on individuals; however, not all individuals have the ability to judge whether information is accurate and credible. Nor does everyone have the time to research and validate such sources. As the volume of information builds, wikis will need to develop information verification systems, or "truth seals," to validate their content. Perhaps these could take the form of more robust ranking and monitoring systems. Perhaps you, as readers and users, have some ideas of your own! If so, I'd like to hear them.

In the case of Wikipedia, the bottom line is simple: A stronger monitoring and review process is needed to ensure the quality of its information. If Wikipedia fails to provide it, the market will. And when that alternative emerges, Wikipedia could find itself stranded on an island with no visitors in sight.


UPDATE: Since the old AlwaysOn site was taken down and posts were not properly transferred, I'm just keeping a copy here.