Forging a partnership between designer and user
Published on September 1, 2004
In the material world, we are generally passive consumers of design. Each day we interact with the elements of our environment—buildings, walkways, vehicles, appliances and so on—and we have very little control over how they look or how they function. Our experience, good or bad, is driven by their design. We do not expect to take control of an element and make it more suitable to our needs or better match our expectations. And so we have a love-hate relationship with design. The designs we love are those we interact with successfully. We chafe at designs that do not fit.
The task of the designer in a fixed medium is to anticipate the needs and expectations of users. It is an impossible task, since all needs and expectations cannot be addressed in a single instantiation, but it’s possible to come close. An iterative, user-centered design process based on a constant dialogue between designers and users can help mold a design to its users. Design based on universal design principles improves usability for everyone by incorporating features that accommodate different needs. But as long as the medium is fixed, there will be people who are disabled by design.
On the Web, we are active design consumers. The Web is a flexible medium, and designers and users share responsibility for its design. The Web designer still must make design decisions and attempt to accommodate user needs and expectations. However, Web users can remove all traces of design and apply their own “skin” to Web content. With so much control in the hands of the user, the task of the Web designer is to design pages that support transformation and leave control of the user environment to the user.
The Web designer
Generally, Web designers take too much responsibility for the user environment. In the early days of the Web, designers felt they needed to make design decisions for its users. The Web was a dismal place for ingesting information because it was visually bereft, difficult to move around in, and generally unpleasant. Designers applied tried-and-true visual design methods to the medium and made the Web better looking and more usable.
Unfortunately, some of those methods are not compatible with the characteristics of the Web. Most of what we know about visual design comes from graphic design, industrial design, and architecture. Generally, these disciplines deal with fixed elements: spaces, structures, objects, and so on. The Web is inherently flexible and users have far more control over the interface than they do over books or stairways or potato peelers. When Web designers apply visual design conventions and then nail them down to ensure their integrity, usability suffers because users cannot modify elements of a fixed design.
The other area designers overstep is in controlling the user environment. The Web behaves in ways that are predictable to users. For example, when a user clicks a link, the browser requests the page from the Web server, the Web server sends the page to the browser, and the Web browser renders the page. Sometimes designers get involved in this transaction by moving the cursor directly to the search input field or opening links in a new window. We, as designers, use these methods because we want to be helpful. We assume that most users will want to use the search feature on arrival; to make things easier, we put the cursor in the search input field. We assume that most users will want to keep in contact with our site while exploring other sites; to make things easier, we open external links in a new window.
But sometimes these helpful interventions wind up causing usability problems because they violate expectations. People expect to begin listening to or tabbing through a Web page from its beginning and will be disoriented if the cursor focus is not at the top of the page. People expect to use the “Back” button to retrace their navigation path and will not be able to return to the originating site if it is not in the window history. While these actions may be helpful to some, they will create usability problems for others. Moving the cursor and opening a new window are functions of the user environment and should be performed by the user.
The Web user
This leads us to the role of the user. For the Web to fulfill its promise of universal access, users must shoulder more responsibility for their environment. There is a wall outside my office window that blocks my view. It is a design element in my environment that chafes. I have no control over or responsibility for the wall so all I do about it is grumble.
On the Web, I have control over (and therefore responsibility for) my environment. If a design element blocks my view of Web content, I can knock it down. I can’t expect the designer to know my preferred text size, column width, and background color. I can’t expect the designer to know that I live in the boonies and only have modem access. I can’t expect the designer to know that I prefer the keyboard to the mouse. But I can expect the designer not to impede my ability to modify my environment and use my preferred methods.
When a designer uses a flexible layout, I can enlarge my text without sending the layout into disarray. When a designer uses images with alternate text, I can turn off image loading and still navigate the Web site. When a designer uses only elements that are keyboard accessible, I can put away my mouse. When given the chance, users can tailor their Web experience to meet their needs and preferences.
To form a successful partnership, both designers and users need to change their approach. First, designers need to think differently about their role. Usability is the primary responsibility of the Web designer—a Web page that cannot be used successfully is worthless. Designers must work within the medium and deliver content in a format that can be transformed as needed by the user. Second, designers need to respect the boundaries of the user-controlled environment. When these boundaries are crossed, even with the best intentions, usability suffers. And finally, users need to become better acquainted with their Web environment. They need to learn the functions and features that are under their control, and use the available tools to customize their environment.
Now, I know what you’re thinking: “That’s fine, but my users can’t manage the ‘Back’ button, never mind customizing their text settings. How can I expect them to master their environment?” While I do not have a blanket solution to raising user proficiency, I do know that users will not even get the chance to control their environment if we keep making decisions for them.
For example, when we redesigned the Web Style Guide site in 2002, we went from a fixed-width, pixel-based layout to a flexible-width, relative-unit layout. Right away we received complaints about long lines (“Text this wide is extremely difficult to read”) and text size (“Please increase the font size of this page” and “The new, larger text is obnoxious”). Initially, we considered reverting to a fixed design, but a fixed design brings its own set of obstacles. A fixed-width column does not adapt to different display devices, and does not accommodate large text. Text set using fixed-size fonts often cannot be resized. And there is a very important difference between the obstacles created by a fixed design and those encountered with our new, flexible layout. When working with a fixed design, user remedies are few to none. It is not easy to extract content from a fixed-layout table, or to change text-sizing attributes to a resizable unit of measure. However, when working with a flexible design, users can easily control line length and text size by resizing the browser window and using the text zoom feature.
Instead of reverting to a fixed design, we opted to forge a partnership with the user. We left the design flexible, but added a section to our Help page explaining how to modify text settings in the browser. Clearly this simple clarification could not have resolved all problems, but we did stop receiving complaints. And by asking users to be more active in defining their experience of our content, we encourage them to take charge more broadly of their Web environment.
Unless we establish this type of partnership, user-centered Web design will remain a benign dictatorship, with the designer making somewhat informed assumptions about what is best for the user. For a truly user-centered Web we need to share control and responsibility with the person who actually knows what is best when it comes to personal usability issues: the user.
Sarah Horton is a web developer with Academic Computing at Dartmouth College, where she helps faculty incorporate technology into their teaching. Together with Patrick Lynch she authored the best-selling Web Style Guide, recently released in its second edition. Sarah regularly writes and speaks on the topic of accessible web design.