Technical trends bode well for KM

What k-loggers do, fundamentally, is narrate the work they do. In an ideal world, everyone does this all the time. The narrative is as useful to the author, who gains clarity through the effort of articulation, as it is to the reader. But in the real-world enterprise, most people don't tend to write these narratives naturally, and the audience is not large enough to inspire them to do it.

Free IT resource Hear how top CIOs turn change into a competitive advantage. Sponsored by HP Free IT resource Attend the SOA Executive Forum: Breaking SOA Bottlenecks SOAExecForum.com/may2007 Sponsored by InfoWorld Knowledge management Test Center Perspective: Products like the Lotus Discovery Server and Tacit'sActiveNet can already find expertise in the enterprise and map relationships by analyzing e-mail and intranet documents. The next step: applying these techniques to voice communication. Business Case: Corporate blogs may be ushering in new era of voluntary communication critical for KM. Knowledge that people don't publish can, alternatively, be deduced from documents and message traffic -- if you take necessary privacy precautions.

There is, however, a certain kind of person who has a special incentive to tell the story of a project: the project managers, who are among the best power users of Traction Software's enterprise Weblogging software, according to Traction co-founders Greg Lloyd and Chris Nuzum (see “Getting Traction”).

"Having a place where that narrative can be recorded, then elaborated by commentary and question and answer, creates a very rich opportunity for people whose job it is to structure the narrative and to be the communication hub," Lloyd says. The Traction toolset makes it easy for project managers to collect e-mail and documents, to inject them into the system, categorize them, and to publish summarized views to intranet Web pages, e-mail subscribers, and the RSS (Rich Site Summary) feeds preferred by a small but growing number of k-loggers.

There will always be a need for automatic ways to propagate awareness within and among teams. With this, knowledge that is not explicitly shared can nevertheless be found and made available. Completing this task, while at the same time respecting individuals’ privacy, is the challenge taken up by Tacit Knowledge Systems. The company's flagship product, KnowledgeMail, builds profiles of people's expertise — and their collaborative relationships — by analyzing the flow of e-mail. It associates people with topics (sets of terms) and then brokers connections among people with specific affinities and those who are deemed expert in that particular field.

To address privacy concerns, KnowledgeMail users control which of these affinities are visible in the public profile, which is encrypted and not readable by anyone, even the system administrator, without the user's permission. Because administrators require read access to the mail store to build profiles, you might think this level of protection is akin to the proverbial "screen door on the submarine," jokes Tacit's CEO David Gilmour. "But we don't see it that way," he says. "When you distill from the e-mail a profile that is much more concentrated and potent, users have greater concerns — as they should."

The same protections also apply to Tacit's forthcoming ActiveNet — which the company describes as a tool-agnostic, "active collaboration network." E-mail messages and other documents such as those in Notes, SharePoint, or Weblogs are fed into ActiveNet and are organized into topical "hotlists" that drive a supply-and-demand market for contacts and information. The topics in your public profile define a supply of expertise in those areas; other users who construct similar profiles create a demand; the system offers to connect those whose profiles intersect.

The Lotus Discovery Server takes a similar approach, leveraging the presence awareness of Lotus Sametime to connect users in real time when possible. Discovery Server differs, however, in its more formal approach to the knowledge taxonomy that's used to categorize documents and affinities.

That kind of upfront investment dooms content-oriented KM systems, Tacit's Gilmour argues. But David Kajmo, Discovery Server's product manager, pushes back, saying, "Rather than bombard you with lists of keywords, we attach well-defined names to the topics we propose." It's best when experts tweak these controlled vocabularies, Kajmo adds, but in many cases a basic taxonomy is already evident in the form of categorized Notes databases or even the naming conventions applied to directories on a file server.

Bottom-up vs. top-down taxonomy is an old, ongoing KM struggle. But the emerging architecture of business process automation may help us cut that Gordian knot. XML documents, produced and consumed by Web services but also by people running a new generation of XML-savvy applications, will be the currency of the information economy. Richly structured, easily captured, and embedded in well-defined business contexts, they'll be a godsend for tools that mine knowledge from documents.