60 Eight Problems That Haven't Changed
Links That Don't Change Color When Visited
Why Designers Don't Believe Us
Breaking the Back Button
Fitts' Law of Click Times
Opening New Browser Windows
The Curse of Maximization
How Can You Use Windows if You Don't Understand Windows?
Most Hated Advertising Techniques
Design Elements That Look Like Advertisements
Violating Web-Wide Conventions
Avoid Influencing Users During Testing
Vaporous Content and Empty Hype
Dense Content and Unscannable Text
84 Technological Change: Its Impact on Usability
1986 Air Force Guidelines Stand the Test of Time
Don Norman's Three Levels of Emotional Design
Slow Download Time
Flash: The Good, the Bad, and the Usable
Low-Relevancy Search Listings
Multimedia and Long Videos
Teenagers: Masters of Technology?
Mobile Devices: A New Argument for Cross-Platform Design?
96 Adaptation: How Users Have Influenced Usability
104 Restraint: How Designers Have Alleviated Usability Problems
Plug-Ins and Bleeding-Edge Technology
3D User Interface
Moving Graphics and Scrolling Text
Custom GUI Widgets
"About Us" Features Don't Say Enough
Not Disclosing Who's Behind Information
Inconsistency Within a Web Site
Premature Requests for Personal Information
119 Assessing the Fate of the Early Findings
The Web has changed dramatically since we began identifying usability problems in 1994. How do our early guidelines from the '90s hold up against evolving technology and Web design? Have changes in users' skill levels, Web sophistication, or expectations had an impact?
This chapter revisits 34 of the most significant usability problems to see which have improved, which are irrelevant, and which are as important now as they ever were.
We conducted our first user studies of Web sites and intranets in 1994. Even the first study, which tested only five sites with three users, identified a number of usability problems. For example, it was clear that users didn't want to read very much; detested long pages with dense, unstructured text; and preferred scannable content. In 1997 we conducted more thorough research into how people read online. It confirmed our early findings and resulted in refined analysis of design flaws and more detailed guidelines for avoiding them. This research was confirmed in yet another project in 2004 and led to even more detailed guidelines for specialized Web writing.
Eight of our original guidelines are as important today as they were when we first identified them. Some bad design practices are actually more of a problem now because continued abuse has made users ever more sensitive to them.
In this chapter, we review 34 of the most prevalent usability issues and our guidelines for addressing them, developed during our initial period of user research, from 1994 to 1999. First, we discuss eight issues that continue to be as critical as ever to usable Web design; then we talk about those that have become less important over time. We have assigned each of the 34 problem areas a skull rating, with up to three skulls indicating how important it continues to be. The rating scheme is as follows:
Still a high-impact usability problem. It is very important that designers continue to pay attention to this problem and that interface evaluations check meticulously for it.
Now a medium-impact usability problem. It is still important that designers avoid making this mistake, but it is no longer a top priority.
Now a minor issue. Designers should remember our guidelines and try to avoid this design error, but it will rarely cause Web sites to fail.
No longer a problem and need not be checked systematically in interface evaluations.
There are three reasons why a usability problem may be less of an issue today, and we discuss each of these in turn:
Technology has improved. When you consider the primitive state of the Web in the 1990s, it is understandable that several approaches caused dramatic usability problems even if they were not inherently bad designs. Improved technology has alleviated some of the difficulties users had in the past.
Users adapt to Web site designs. Confusing designs create many usability problems. When people get accustomed to certain design approaches, however, they are no longer confused, and those design flaws are less of a problem.
Designers refrain from the worst abuses. To the extent that designers have shown increased understanding of particular design flaws and restraint in using approaches that increase the potential for problems, we consider that element to be less important, and we don't need to warn against it as strongly anymore.
Unfortunately, we can't revisit many of our minor findings in detail here. For example, in 1994 we found that it was a bad idea to post "under construction" signs on Web sites. Users said, "Either you have it or you don't," and they didn't want to be teased with links to material that was not yet available.
The main guideline for under-construction signs is to avoid them and not advertise features until you have them. If this is not possible, at least provide an estimate of when the information will be available. Even better, if it is an important feature or a product that you will be selling, offer users the option of receiving an email announcement when the page goes live.
Even though "under construction" signs are not nearly as common as they were in the 1990s, we still see them from time to time. For example, in a recent consulting project for a big company's intranet, we followed a link for a useful feature, only to land on a page that read, "coming soon." Our early finding certainly remains true todayusers still don't like clicking in vainbut we don't emphasize it anymore because under-construction signs have become so rare.