Karen Holtzblatt, Jessamyn Burns Wendell, and Shelley Wood
Morgan Kaufmann ISBN 0123540518 $39.95
Rapid Contextual Design: A How-To Guide to Key Techniques for User-Centered Design is a successor to Hugh Beyer and Karen Holtzblatt's 1998 book Contextual Design: Defining Customer-Centered Systems. That book documented a methodology for extracting design requirements from users and how to package them in a way that could be readily used by all divisions of a product-development organization, not just the HCI folk. Some found it difficult to read and repetitive, but most agreed that the content of this original book was valuable.
Rapid Contextual Design is described as a companion to the original work, rather than a replacement for it. The purpose of this companion is to offer practical examples and tips and to show how the contextual design (CD) process can be tailored for use when time frames and resources are limited. It is intended to be a book of practice; the authors recommend that readers turn to the original book for a "deeper discussion of the method and philosophy" of CD.
Useful Content. So, how well do the authors succeed in providing CD practitioners with a more rapid and tailored approach? The word "rapid" in the title is obviously a reaction to the constant industry pressure to reduce development time, and the main aim of Rapid CD is to help practitioners to be more flexible. Rather than proposing ways for speeding up the essential parts of the CD process, Rapid CD instead helps practitioners to select the key parts of the CD process and leave out others so that a study can be completed in a shorter timeframe. The authors identify three speed-settings for Rapid CD: Lightning Fast, Lightning Fast +, and Focused Rapid. They do a good job of describing how to perform each of the three types of studies, but the terms they have chosen are unfortunate. It wasn't immediately clear to this reader which type of study was fastest (is "Lightning Fast" faster or slower than "Lightning Fast +"?) and which was the most thorough.
The authors also propose a few useful changes to the original CD methodology. One valuable new addition is the section detailing how to use CD data to help in writing Personas. Their treatment includes a number of samples, tips on how to use Personas successfully, and some CD-specific extensions to the technique.
A successful feature of this book is the inclusion of a number of real-life projects as instructional samples. One project in particular is used throughout the book as a consistent touchstone for the reader. All of the examples are extremely useful as indicators of how to take a pragmatic approach to a process that may at the outset feel overwhelming.
Some of the most valuable content in this book is its many tips on how to perform CD studies. These tips, obviously gained from hard experience, help to ensure that readers learn from the authors' mistakes. They are all about the practice of CD, and so are equally applicable to teams working on software applications, manufacturing processes, or consumer products. There are tips on how to dress ("one level of formality" up from your users), why interviewers should not go out in groups of two (the second for taking notes can get in the way), how study notes should be taken (don't use a laptop, please), and the logistics that should be taken into account (always confirm appointments the day before, and arrive on site early to deal with security and other visitor processes). The value of such practical advice cannot be underestimated. Even if you choose to ignore some elements of the authors' advice, you will at least be aware of some potential pitfalls to avoid.
Haphazard Editing. The main criticisms of the book are not in the domain of the authors, but of the editors. It is large and overlong, poorly formatted, and poorly edited. For such a large and complex book, one would expect a robust content structure and organization, typographic and layout details to signal this structure, and simple and consistent language to make the extended arguments clear and easy to follow. If this book is to be a handbook or guide, it should be formatted as a reference work and not as a linear dialogue, to help users "dip in" for specific information when they haven't read the surrounding chapters recently. This book was not well-executed on both of these features.
The first thing that CD aficionados will notice is that the Rapid CD book is both wider and taller than its companion (the original CD book). In fact, it uses 8.5" x 11" paper. This format hardly befits a companion, as this new book may not even fit on your shelf beside the original work. One would also expect that a companion handbook would be somewhat shorter than the original work. That is not the case here. Even though Rapid CD has fewer pages than the original, they are larger, and so the handbook is only seven percent smaller (by paper area) than the original.
It is possible that this new format was used with the good intent of presenting larger figures and diagrams, and large two-page spreads. There are quite a few large figures, but unfortunately no large graphic spreads. The editors did not capitalize on the advantages of using this larger format. In the end, the use of larger paper just seems to have turned into a license for presenting more content than might be recommended for a handbook.
In terms of a clearly signalled structure, this book again differs from the original CD book. The original included 20 chapters wisely grouped into seven sections; this simplified the structure and use of the book. This new book has 16 ungrouped chapters; the lack of obvious grouping makes it difficult to discern the architecture of the book as a whole. Adding to this, sometimes there are only minor typographical differences between different heading levels. This only further disorients the reader.
The main weakness of the original book is that it was difficult to read and repetitivea weakness that is not corrected in the companion. Presumably in an attempt to be collegial and approachable, the authors use an informal style of prose. Unfortunately, the book can be a little too informal, and uses words and sentence constructions that are sloppy and militate against conceptual clarity. As an example, one paragraph (page 159) begins, "The affinity diagram is relevant for any form of Rapid CD. It is your fastest and best method to see all the issues across your user population." Then, in the space of six lines, we also read about "whole market," "whole population," "population as a whole," "whole user population," and "not just one individual." The terms "population," "market," and "not just one individual" could reasonably have different meanings, but perhaps not. Who's to tell? In the end, we're left with what might or might not be a big conceptual muddle.
Recommendation. The original CD book by Beyer and Holtzblatt made a compelling argument for a model process to capture valid data from real users in a way that would empower real software engineers to make real product improvements. It was a valuable and practical help to the emerging practice of user-centered design. This companion guide provides valuable and very practical material to support CD, and can help the uninitiated to tailor the process to meet cost, resource, and time constraints.
The faults of the book are many, but are mainly due to poor editing. The larger format is more of a hindrance than a help, the high-level structure of the book is not obvious, and the typography and layout make it difficult to find the many real nuggets in the text. The editors should have been more aggressive in their control of informal prose, excessive repetition, terminological confusion, and item ordering. If the book was half the length and better organized for easy reference, it could have been an essential reference on the shelves of all user interface designers. As it is, it can be a valuable asset only to those willing to see past its flaws.
About the Reviewer:
Tim Moore was a founding member of the Human Sciences and Advanced technology (HUSAT) research group at Loughborough University in England, and worked on human computer interfaces almost a decade before the first personal computers. For five years Tim was the prime HCI consultant for L.M. Ericsson, the Swedish telecommunications giant. After obtaining his doctorate, Tim moved to Canada where he worked at Nortel Networks, developing advanced telephones and for a period managing the then newly emerging Usability Laboratory. He has recently worked on Avionics systems in the Human Factors team at CMC Electronics. He now runs Ergosum Ltd., a Human Factors and Ergonomics consulting company.
To submit a book review, please email Gerard Torenvliet at firstname.lastname@example.org
415 Leggett Drive, P.O. Box 13330
Ottawa, Ontario, Canada K2K 2B2
©2005 ACM 1072-5220/05/0900 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.