Categories

Categories, representing the subject matter view of the content in your workspace, differ from the taxonomy structures addressed by the folder adaptation method. As stated before,

• Categories are reader-focused.
• Categories follow no security policy restrictions.
• Documents in the workspace can be placed in multiple categories.

The readers who use categories could be the content authors who are very familiar with the workspace, readers who have a basic view of the workspace without intimate knowledge, or casual exploring visitors who have no knowledge of the workspace structure or content. These latter groups most need the browsing guidance that categories provide, so they should remain the primary focus when designing category structures.

As stated earlier, the folder hierarchy is not the optimal basis from which to derive categories. As the folder adaptation method shows, the concerns that drive folders are not the concerns that drive readers. Categories should adopt an external view of the information in the workspace. One way to gain an external view is to look at materials designed for external consumption. Another way is to listen to the questions people outside the organization ask about the material in the workspace, and design categories that guide those people to the answers.

As an example, one team reviewed marketing presentations and external reviews about their projects to understand how to organize information for consumption outside of the team. Because people whose expertise lies in communicating project concepts to unfamiliar audiences had built the presentations, using the materials leveraged the work that they had already done. Then they used the information breakdown from the presentations to guide the formation of the initial set of categories.

Unlike folders, where it should be as easy as possible to identify the single location where a document belongs, categories should not shy away from redundancy. Multiple categorizations allow the category designer to anticipate and respond to the many ways that different readers search for information. Having multiple category paths to a given document increases the odds that the reader locates a document of interest. For example, a reader searching for content about the U.S. Congress might start with Government, then Federal Government as opposed to state or local, then Congress, while another reader might start with National Legislatures, then U.S. Congress as opposed to the Japanese Diet, Israeli Knesset, or Russian Duma.

Every level of the category tree can point to content. Beware of intermediate categories that provide no value and force users to browse through before they can find anything. Each intermediate category represents additional layers of navigation standing between readers and the content they seek. At worst, they increase the complexity of navigation through the category tree. In addition, although you can organize information into categories to an arbitrary depth, excessive category depths challenge readers. Any category structure deeper than three or four levels merits review. You should limit the maximum depth to that with which your readers are comfortable.

The categories defined on the first day of the workspace design do not remain static. Expect that the categories evolve over time in response to the needs of the users. Beyond responding to specific requests from users, you can explicitly capture the real needs of users by monitoring the most frequent searches, and striving to turn those search terms into categories.