March 10, 2002
Critical note: Hackos, J. T., and Redish, J. (1998). User and Task Analysis for Interface Design. New York : John Wiley & Sons.
This work is aimed at the designer or consultant who wishes to evaluate interface design in the workplace. The authors provide a thorough plan for analyzing tasks and users, taking the reader through each step from how to define users to prototyping for interface design after gathering data. They cover a wide range of situations, with numerous brief case studies to illustrate methods. The authors note in the preface that they have written to an audience of software designers, interface designers, usability specialists, human factors specialist, instructional designers, or technical communicators, and point these readers toward specific chapters. As such, it covers a much wider approach than my goal of analyzing website use and usability within my own workplace, but as the authors promise, many elements are directly applicable.
The key concept for the authors is the importance of understanding what users do with a task or product. They blame unusable designs on designers who have not worked with users to develop products that make tasks easier, rather than more cumbersome. Good interfaces will be nearly transparent to the user, fitting with the users' life and work needs, with certain characteristics in common (p. 7):
Several chapters focus on conducting site visits, and in particular, on honing observation and interviewing skills. My project doesn't lend itself well to observations, but, since I have good evidence that colleagues will be unlikely to respond to surveys, interviews will be one of my primary means of gathering data. The authors emphasize a familiar concept to anyone with reference experience - neutral questioning. They offer a taxonomic chart of question types that would be useful in both usability and library settings, with additional examples of each type. They also give tips on eliciting additional information from users through paraphrasing, prompting and direct questioning.
The authors recommend prototyping as a useful means of testing designs. They discuss a range of options, from sketches on paper to high-fidelity computer prototypes that let users work with a product in its early stages. They note that prototyping "saves money, is concrete, fosters alternatives and iterations, and allows users to be actively involved at the design stage" (p. 379). In my own experience, prototyping is a critical step, because the picture really is worth a thousand words; poor design becomes much easier to avoid when a variety of people can look it over before production. They do offer the important caveat that demos and reviews of prototypes are no substitute for usability testing. A flashy design will not meet users' needs if it doesn't make the job easier.
The best feature of this book is its emphasis on testing designs and ideas with the actual users, rather than management, a select group of expert users, or worse - with no consultation at all. Due to time constraints, my project website was developed with direction and feedback from a few managers and no staff involvement. As a result, a variety of problems exist due to no usability studies before putting it into production. As the authors note, "designers who spend time with users, observing how they work, understanding who they are, testing concept designs and prototypes, are most likely to be successful in creating interfaces that are a delight to use" (p. 7).