The book on Web design

Take a team of experienced Web design researchers, add the evidence of hundreds of research studies, pour it all into a user-friendly format and stir. The result, 'Research-Based Web Design and Usability Guidelines,' is a handy 267-page cookbook for building better Web sites, produced jointly by the Health and Human Services Department and the General Services Administration.Based on hundreds of research studies in areas such as cognitive psychology, visual comprehension and search technology, the handbook is the only design and usability guide to offer Web design tips founded on empirical research, not opinions or guesses.Author Sanjay Koyani, director of Web communications for HHS' Food and Drug Administration, began working on the book in early 2000. The guidebook was jointly authored by Koyani and a team of 17 other researchers, engineers and consultants from government agencies, academia and industry. The first edition came out in 2004; the second edition was published last year.'The rationale for the book was that we were seeing a lot of common mistakes, both in and out of government, in Web design,' Koyani said. There were a few companies that published guidelines for Web design: Microsoft, Sun Microsystems, IBM. 'But we found no research tied to any of it.' Such guidelines were too often based on a 'this is the way we've always done it' approach, Koyani said.HHS is a research-based agency, Koyani said, so it wasn't surprising that it was one of the leaders of the project. 'We wanted to make sure our guidance was based on research, not opinion.'The group looked at all the major research available on every aspect of Web design and usability. 'A thousand research studies go out every year,' he said, 'but most of us can't figure out what the devil they're saying.' The group spent more than a year translating the research into plain language and practical guidance. The book limits each guideline to one page or less and shows both good and bad examples to illustrate the design and usability principles.[IMGCAP(1)] One reason the two agencies partnered is because GSA runs USA.gov, the official portal for the federal government. 'We believe USA.gov is only as good as every other government Web site, because we link to every other government Web site,' said Bev Godwin, director of USA.gov and Web-based practices for GSA.The group came up with a visual shorthand to show the relative importance of each guideline and also how much research had been done on it. Each researcher ranked the guideline on its relative importance and how strong the research on it was on a scale of 1 to 5, with 5 being the strongest. Each guideline is accompanied by two sets of five circles. The top set shows the relative importance of the guideline. The lower set shows the strength of evidence about the guideline. For example, Guideline 2:2, 'Do not display unsolicited windows or graphics,' rates a relative importance of five circles; its strength-of-evidence rating, however, is only three circles.'So if a guideline has only one circle indicating its relative importance, don't fall on your sword for it,' Koyani said. 'Or if the research isn't so strong, you might have to do some of your own research.'The government took the lead on the guide, in part, 'because nobody else was,' Koyani said. Most of the research was performed by universities and private industry, but the government had the resources to compile the information relatively quickly and inexpensively, Koyani said.The back of the book lists each guideline from most important to least important, and also ranked by strongest evidence to weakest evidence. 'That way people can see at a glance what are the most important guidelines,' Koyani said.As the Web evolves and the agreed-upon conventions of the Web change, so too will the guide be updated. Koyani estimates the book will be updated in two-year cycles.[IMGCAP(2)] Even though the Web is a relatively new medium, certain conventions have emerged, which the guide codifies. For example, underlined blue text indicates a clickable link. Web designers should avoid using underlining for emphasis, the way you would in print text. Or the 'mousing over feature,' where the movements of a user's mouse reveal additional text, tends to be more confusing than a straight point-and-click feature.The guide uses examples from all over the Web, not just from government sites, although the majority of examples, both good and bad, come from government sites. CNN.com, for example, is used to show effective use of attention-attracting features when appropriate.Some of the guidelines seem obvious but worth repeating. For example, guideline 1:1, 'Provide useful content,' probably can't be overstated. The relative importance and strength of evidence both rate top marks of five circles. Content is king. Bells and whistles are fine, but great packaging without solid content is a waste of resources.Ben Shneiderman ' professor of computer science at the University of Maryland ' who wrote the forward for the book, calls this the best of the guideline books for informational Web site design. 'Ensuring that there was empirical evidence for each guideline was a massive effort that the authors did diligently,' Shneiderman said. 'The charm of these guidelines is that the value and benefits go far beyond government Web sites.'With little in the way of an advertising budget, the guide has held its own in the world of government publishing. The book is being translated into Japanese, Chinese, Italian and German. The Government Printing Office sold out of the first printing of 4,000 in three months. Every student who signs up for a day-long class at GSA's Web Manager University also receives a copy of the book.'Research-Based Web Design and Usability Guidelines' is available for free as a PDF at www.usability.gov, or as a bound paperback for $25 from the Government Printing Office (www.gpo.gov).