SUBSCRIBE NOW!

Vol. 27 No. 2 — March/April 2003

Are we still in the early days of online information? Of course, digital databases have been available for decades. Most people know how to pull up a Web page and type a few keywords into Google. But finding specific, high-quality information efficiently still is a game of skill, and complex searches often require a trial-and-error approach—you still need to "fiddle with the dials."

Content providers that want to make working with online information easier can follow usability guidelines to design their systems and services, but what exactly does usability mean now that we are a few decades into the digital information revolution? Are there different guidelines for different types of content? Is there anything new in the field? And are there emerging technologies for which usability principles need to be applied?

DEFINING USABILITY

Usability is a relatively new field, but the term has been used so often in so many different contexts, it is in danger of losing its precise meaning. Andrew Dillon, dean of the Graduate School of Library and Information Science at the University of Texas, says he tends to "accept the ISO definition of usability which refers to the effectiveness, efficiency, and satisfaction with which intended users can perform realistic tasks in context."

According to Jakob Nielsen, principal of the Nielsen Norman Group, "The definition of usability is to help humans overcome technology and make it easy and efficient and pleasant for them to use."

USABILITY DEPENDS ON CONTEXT

But how are those definitions applied in different contexts? Is usability different if you're talking about a library catalog, a library portal, a library site in an academic setting, or a library site in a corporate setting?

Dillon points out that "usability is a property of interaction, not of interfaces. So usability transcends portals, settings, and catalogs. However, in determining and defining usability for each of these contexts, it is reasonable to assume there might be some differences in the desirable levels of effectiveness, efficiency, and satisfaction we wish to enable. Given this, the resulting usable interfaces may look very different for each of these scenarios."

Nielsen says, "Fundamentally, usability is always the same, but the interpretation of the words in the definition can be very different. In an academic environment, for example, there may be less need for an easy-to-learn system because it's the user's job to learn. Students are supposed to learn, and professors are supposed to do research. So those users typically will be more interested in something that's a very powerful tool that can really help them retrieve all the information they need. Also in academia, there's a premium placed on having done thorough research and being familiar with the literature, so those are the tasks you would facilitate in that setting.

"In the corporate setting, it's like, 'Don't bother me with that stuff,' but it depends [on the specific environment], because obviously there are multiple types of corporate settings. There are corporate research labs that are similar to academic environments, so the distinction would be 'Are you doing research, or are you doing business?'"

If you're doing business, Nielsen says, the traditional information retrieval system—one that provides the ability to find all relevant information on topic—often is not appropriate.

"That's not really what you want in the business setting because you don't have time to study all the information," he says. "You just need a summary. You need the facts—the one best answer, not all answers. In that setting, usability means efficiency. How fast can a user find something? In addition, you cannot necessarily assume the average businessperson is going to care very much about library catalogs or anything like that. The technology has to be really simple."

WEB DESIGN

So what do those guidelines mean for Web design? How would the design elements differ if you're providing, say, a directory of corporate information, a bibliographic database, or a numeric file?

According to Dillon, the differences are best treated "through an analysis of the intended users and their tasks."

Nielsen agrees. "You want to present information so it makes sense for the tasks people are trying to perform," he says, "because the types of information are quite different, and you're going to want to do be able to do different things with them. For example, you may want to be able to sort one type of data by different criteria. I think the most popular way to do sorting is to present a table. Then the user can click on the table to sort by heading. It's an idea borrowed from spreadsheets that we have generalized into other systems. This holds to the principle of consistency: Once I've learned one way of doing sorting, if I can use the same way in other interfaces, it's easier.

"But on the other hand, because there also are going to be a lot of differences on those types of sites, the way you would resolve this question is to ask yourself, 'What are the top three most important things people are going to do with this information?' Then you're going to emphasize those tasks. Of course, you look at the other things people are going to do, and you would try to make those tasks possible as well, but you always want to emphasize the top ones."

A NEED FOR CREATIVITY

Dillon says he believes usability is hampered by a lack of creative approaches to enabling information-related tasks.

"Most interface designers are stuck in a groove," he says, "and there is a dearth of really creative new design occurring. This is most apparent on the Web, where the standard navigational bar layout and the unimaginative exploitation of spatial properties have resulted in a sameness to most sites.

"I suppose the rigid adherence to rather vague heuristics as if they were laws has not helped, but we really need to start questioning what works and why it works from a human perspective so we can produce designs that truly augment human capabilities."

Dillon adds that although "this is the age of augmentation and empowerment, most usability work tends to trivialize these concerns."

WHAT'S NEW IN USABILITY?

During the past decade, studies have revealed new insights into the best ways to apply specific usability guidelines, but Nielsen points out that "the basics of usability—making technology easy and efficient and pleasant—have been the same forever, and the bigger conclusions as to how do we achieve usability have been the same too. Consider the principle of consistency mentioned earlier. If you have to learn things only once and then you can do them the same way everywhere, that dramatically lowers the learning barrier. And that's fundamental. That's not dependent on the specific technology we're using today versus 10 years ago. It's not dependent on whether we're using mainframes or PCs or Web sites or cell phones.

"Or," Nielsen continues, "consider progressive disclosure, which means you show a smaller thing first and then show a larger thing. That's a fundamental principle as well. Of course, the interpretation of these principles can differ. For example, in a traditional PC environment, an example of progressive disclosure is to show people a dialog box that contains the most important choices—'how many copies do you want to print,' for example—and there would be a button that says something like 'advanced features' or 'advanced print options.' You would click that to get the rest of the choices. For Web navigation, you often like to have, say, a summary of articles on the first page. Then you can click to get the full articles. It's the same idea, but the details are certainly different for different environments.

"Still, the basics of usability are the same, and the reason they are the same is they relate to the way humans work. That doesn't change over a 10-year period. It's literally the same people. I'm here today. I was here 10 years ago. I'm going to be here, hopefully, in 10 years as well. So it's the same literal warm bodies we're designing for year by year."

NOT JUST WEB SITE USABILITY

Nielsen points out that during the next few years, applying usability principles will be important not only for designers of emerging technologies, but also for the information organizations that will be providing content to the new devices.

"I think we are only at the very beginning of where we're going in terms of technology," he says. "I think the biggest change is probably going to be liberating the Web from the PC, once we get much more portable interfaces."

"Portable could mean small handhelds that you carry at all times, or it could mean flat-panel tablets that give you, for example, the ability to read the newspaper in the morning at your dining table. I think we're going to get a new generation of portable devices, and that's going to generate a vast outreach in terms of additional things we can do that we can't do right now. That's something people should look at. They should ask themselves, 'What does that mean to what I'm doing?'"

Perhaps in the future, people will look back at this decade and our struggles to get Web site usability designs right, and marvel at how primitive our interactivity with information resources really was. They may not liken it to "fiddling with the radio dials"—a different metaphor may come to the fore. It's likely, however, that ease of use, the ability to access the correct information quickly, and customer satisfaction will remain as important elements of the information retrieval experience.