Reading Usable Help
@UsableHelp on Twitter
Gordon R. Meyer
The Ghost In Your Machine, by Chris Forsythe for Business Week, examines the possibilities of "smart computers" that would remember your past experiences, alert you to mistakes, and adapt to how you use your system.
The application of this idea in regards to onscreen Help is fascinating. Writers are often required to create documentation for the widest possible audiences, resulting in a "one size fits none" approach. But if the help system could adapt itself, based on the users interests and usage patterns then the result would be a more dynamic and personalized experience.
The first step is to develop usage models for various types of customers. One way to accomplish this is to enable current help systems to gather data about how they're used. When Help is hosted on the Internet, server logs provide some clues about the pages users are looking at, but their motivation and how they found the page (searching, browsing, or a direct link) are all missing. Using cookies and IP address tracking would add more info, but asking customers to submit their search terms, qualitative assessment of the help, and capturing the context of what they were trying to accomplish would form the first steps towards creating a "smart" help system.