Knowledge management, aka “KM,” is, ironically, responsible for much confusion. Little wonder it’s so tough to understand what it’s all about when even the buzzword KM has its own buzzwords. A few years back, “taxonomy” was the KM buzzword. A taxonomy, the thinking went, was the holy grail for a law firm seeking to leverage its knowledge. The firm could classify its knowledge, creating categories for every practice area, every subject, every jurisdiction. In short, once the classifications were completed, the firm would have a total map of the kinds of things it knew, making the sorting and retrieval of that know-ledge a simple process.
But there’s a challenge inherent in taxonomies—the requirement of thoroughly classifying everything—which is precisely why so few law firms have been successful in implementing them. An effective taxonomy requires tremendous commitment on the part of the entire firm, not just the IT staff. To be useful, it must reflect what the lawyers know and, therefore, requires the time of the lawyers—the same ones who have little time to spend on nonbillable work (especially when it concerns technology).
Now a new concept is emerging on the Internet that may lead to a more workable approach to sharing knowledge. It’s a tool not designed to solve law firms’ KM needs (a market rife with expensive products and law firm budgets in six and seven figures), but to solve … a bookmarking problem? Yes, as in your browser bookmarks.
Social Bookmarks and the Basics of Tag
In his book The Innovator’s Dilemma, Harvard Business School professor Clayton Christensen describes how many new technologies develop: by delivering less functionality than customers are really asking for. (Doesn’t make much sense, does it?) Looking at a number of technologies that developed over the 1980s and ’90s, Christensen pointed out that in each case, the newer technology was not what the customers were asking for but instead often came out of a tangential business need. By serving the lower (and less expensive) end of the market, these so-called disruptive technologies grew and eventually represented a threat to the established companies.
Among similar technologies in play today, we have what are known as social bookmarking sites. Such sites work as a “social system” because users can see the links that others have collected, as well as who else has bookmarked specific sites, so you can subscribe to the links of people whose lists you find interesting.
To illustrate, visitors to the social bookmarks site http://del.icio.us (a service far easier to pronounce than to type) can create a free account, where they can start bookmarking the Web sites they visit. A bookmarklet provided by del.icio.us lets you easily add sites you like to your personal collection of links and categorize those sites with keywords. Whenever you bookmark a site, del.icio.us prompts you for some words to describe the site. Each word becomes a “tag”—and each tag becomes its own category that lets users navigate through the bookmarks.
Unlike the taxonomy approach, there is no relationship—hierarchical or otherwise—among tags. Actually, once you realize that del.icio.us is designed to be a social service, you start to see there’s an opportunity for chaos: You have no monopoly over what tags describe a particular Web site.
Take a CNN.com story about the weather in Chicago as an example. Once I see it, I may add the tags “chicago, weather and snow,” while someone else may add the tags
“illinois, accumulation and windy.” In a traditional, top-down view of knowledge management, this is fundamentally broken: Either the article in question is about “Chicago, weather and snow” or “Illinois, accumulation and windy.” Or a combination of the two. But not either one or the other. There has to be one, agreed-on view of what the CNN article is about (and what it means), or there will be uncertainty and confusion.
Tagging as a Window into Your Colleagues’ World
Except that it turns out there isn’t uncertainty and confusion. When I use del.icio.us to look at sites I’ve bookmarked, I can see how I categorized the site, as well as how others have done it. In this sense, I get to see angles to the site’s contents that I may not have realized existed. I see alternative perspectives about what the site means to others. And, thanks to the magic of hypertext, I can navigate through those alternative tags to view other sites that are similarly categorized.
Rather than force a group to agree in advance on what terms matter, a service like del.icio.us lets the group instead focus on which content matters. The categories exist as representations of the content, rather than the other way around. Broad tags like “Chicago” can include weather, restaurants, real estate—anything. And narrow tags like “accumulation” will ordinarily contain targeted information specific to that concept.
In both cases, the possibility of finding things you didn’t already know about is quite high. Since the goal of any good KM collaboration strategy is to spread the organization’s knowledge among all participants, this is a good thing!
This bottom-up way of adding order to the content (not vice versa) proves to be a very effective process for not only classifying the content, but also for identifying which categories are, in fact, interesting to the group.
This site and similar ones are, admittedly, consumer services. But let’s return to Christensen’s hypothesis in The Innovator’s Dilemma–that disruptive technology often comes from the lower end of the market.
As of this writing, a major endorsement of the tagging phenomenon comes from Technorati, a company that grew up around indexing RSS feeds from weblogs.
Technorati Tags: Exposing Connections
Technorati ( www.technorati.com) is a service that shows the linkages among “conversations” on the Internet. Weblog A, for example, links to weblog B, which links to weblogs C and D.
Those links represent connections that Technorati indexes, then represents on its Web site. Anyone visiting Technorati can thus find blogs that are related to each other. Let’s say you like Ernie the Attorney’s blog ( www.ernietheattorney.net), but didn’t know about Bag & Baggage ( www.bagandbaggage.com), the excellent blog from attorney
Denise Howell. Thanks to Technorati, all you’d need to know was the Web address for the Ernie the Attorney site; once that’s plugged into Technorati, it will reveal all blogs that link to the first site. This alone is useful.
However, a new feature from Technorati now also indexes tags. As a result, rather than just showing you links between Web sites, Technorati can also show you the links between tags on various sites. This allows you to navigate from one blog to another based on the category the blog author chose to apply to the content. Technorati also indexes tags from del.icio.us and various other social sites, like the photo-sharing service flickr.com and Socialtext wikis.
When you can follow the linkages among content in this way, suddenly the lack of a hierarchy is no impediment to finding related kinds of content regardless of where that content is stored or who categorized it. (And unlike traditional systems, where the author dictates how the content is categorized, sites like del.icio.us let each individual viewing the page decide how to categorize the content, increasing the likelihood that subsequent visitors will see it.)
To fully understand why the concept of tagging matters, and what it means for how individuals and organizations will use the Web in the near future, I recommend David Weinberger’s Small Pieces, Loosely Joined: A Unified Theory of the Web (Perseus Books, 2003). It recounts Weinberger’s experience with the competitive displacement of the Web browser in 1993, and how the Web is remodeling our concept of society. (You can read the provocative preface to Weinberger’s book at www.smallpieces.com/content/preface.html.)
A Folksier Approach to Commerce
And finally, back to the topic of buzzwords.There’s even a buzzword for the phenomenon discussed in this column: It’s folksonomy—which Wikipedia defines as “a practice of collaborative categorization using freely chosen keywords.” You’ll find the complete Wikipedia definition and a list of resources at en.wikipedia.org/wiki/folksonomy. I’d like to think we’ll see enterprise applications—that is to say, commercially produced software based on these concepts targeted at businesses—within the next six months. One thing is certain: Whether I’m right or not about all this, the tags that others use to describe this article out there in the blogosphere will tell the story better than I could.