Wiki goes to work

In last week’s column, I talked about the potential of Wiki as an enterprise tool and promised more details on that topic this week. A recent conversation with Peter Theony, the developer of TWiki, got me thinking about how to solve some specific knowledge management problems I encounter in my daily work as a CTO. Although Wikis are still largely an underground IT phenomenon, a Wiki can be one of the most immediately useful tools in an agile IT organization.

Wikis are organized as topic-based Web pages that are editable by anyone. In my organization, I might have a page describing changes to our Apache Web server configuration, a page detailing firewall rules for a WAN connection, and another page recording any hardware maintenance done to servers. As changes are made to specific pieces of the IT environment, IT staffers can use a Wiki page to record them as they happen.

The word “wiki wiki” is Hawaiian for “quick,” and the quickness with which Wiki pages can be edited and searched means it’s easy for IT staff to record important information and find it later -- easier even than it is with blogs, and far easier than conventional groupware solutions such as Lotus Notes. If the ease of editing and changing seems scary, then it’s worth noting that many Wiki systems, TWiki included, can be configured to track changes with revision control. Ease-of-use doesn’t sacrifice accountability.

After digging into TWiki, I came up with two simple uses of the software right away. One of the unique features of TWiki is its forms capability, which allows users to create quick-and-dirty, structured, databaselike applications. Using TWikiForms, I was able to design a simple workflow system for rolling out laptops to end-users, built around a simple TWiki-powered, Web-based checklist our IT staff could use to ensure that key functions such as VPN access had been tested before being delivered to end-users. A simpler use of the TWiki I devised was setting up a page for each end-user as a quick trouble-ticketing system. On those pages, we keep problem notes that allow us to quickly discern whether a user is having a recurring problem or needs training.

Because Wikis and blogs are often mentioned in the same breath, it’s important to point out a key distinction: Blogs are an excellent tool for knowledge management in IT, but in practice, blogs have shortcomings. One of the hallmarks of blog structure -- a reverse chronological listing of posts -- can actually be a drawback for certain types of information. In some contexts, instead of having the newest information at the top of a page, you might want the most important information at the top.

Suppose you post to an internal blog to document how to fix a particularly thorny (but commonly occurring) problem with one of your key systems. Right after you, a few of your colleagues add to the blog with some minor issues that are worth noting but are not incredibly important. In a blog, the important post gets bumped down the page by the newer ones. In a Wiki-based knowledge management system, the lack of structure means that information can be arranged in any way that makes sense without restriction. You can leave the most important information at the top or anywhere else.

Blogs can be quite useful for behind-the-firewall knowledge management, but if blogs seem too rigid, take a look at Wikis. In this case, “quick” is truth in advertising.