1.Usability is a buzzword in the field of Web design. Unlike most buzzwords, usability will always be a vital component of design for the Web, particularly because usability is a core requirement of any computer-human-interface design. The root of "usability" is "use" and the obviously related noun of agency is "user." Rhetoricians and designers alike have placed the user in various positions relative to the design process, from design based in Control of user experience and options to user-centered design a la Robert R. Johnson. Rather than position these two as a binary in opposition, one may better conceive of a spectrum within which designers must operate, with Control at a user-devoid extreme and Consensus at another, more "personalizable" extreme. The intervening grey area is what the designer must navigate, with the components of time and resources as primary filters and lenses for decision-making. Another useful lens is culture (or community).

2.As the above quote from Stephenson points out, graphical user interfaces are not always the best tools for accomplishing all tasks. Unfortunately for Web designers, the GUI is one of very few options for interaction. Though browsers like Lynx are still in use, and though access to Web sites via text-based, streamlined browsers on devices like PDAs and cellphones is becoming more commonplace, the Web as a cultural institution in the United States and most of the rest of the world is primarily a visual medium with a point-and-click interface driven by a mouse. Many users don't know keyboard shortcuts within their browsers, and some are not even aware that hitting the "return" or "enter" button will complete a form or URL. Everyone clicks.

3.The main element of the Web, though, from a content perspective, is text. People read on the Web, or seek textual information. Images form another large portion of content. What may be overlooked, and often conflated, is the use of text and images as both content and interface. No other controls exist in Web sites. Links are visually enhanced text, and "buttons" are images that trigger links and register that triggering by reacting with an altered image. There are no levers or pulleys or steering wheels in cyberspace. Integrating such devices via external hardware accessories is an exception, and a very slight one, and not the norm.

4.Given these limitations, how then is a designer to accomplish the task of building an interface that makes sense to users--that is "usable" with only a GUI or textual interface? Jakob Nielsen suggests avoiding most images, using text and very little of it, and not deviating from default settings as far as link colors. His approach includes the "Top Ten Mistakes in Web Design," a list of technologies Nielsen fervently recommends designers avoid. Three years later, he revisits these "Mistakes" and suggests that most are still problems. In the years that follow (1999, 2002, 2003), Nielsen describes a sort of "year's worst" mistakes, rather than revisiting his original ten. His analyses and recommendations are usually savvy and helpful. But they are only recommendations.

5.Julie Meloni takes Nielsen to task. In "Tipping Jakob's Ladder," she points out the tensions between usability, aesthetic, and function and acknowledges the need for guidelines like those set forth by Nielsen, even suggesting that [e]very site designer should be forced to read [Nielsen's usability guidelines]. She suggests that young developers seeking to create the next 'killer app” are precisely the people who need a smack upside the head with a stack of usability rules. But Meloni also points out that Nielsen's very own UseIt.com only got a C for usability.

6.Designers need to treat guidelines as such, and not as commandments, no matter what Jakob Nielsen says. By learning the guidelines, one can break them with a purpose, for a reason, with a goal in mind. Another mistake Meloni tasks Nielsen about is the notion that a technology is bad or should be avoided entirely. Meloni rightly critiques Nielsen's strategy as a sort of lowest common denominator approach that wrongly blames technology for mistakes made by designers. Nielsen's approach also asserts that Web users do not scroll, and he explores in one article "Why Web Users Scan Instead of Read." Meloni, Derek Powazek, and other usability designers offer compelling rebuttals.

7.Donald Norman, a colleague of Nielsen's, also writes about usability. His work is useful for designers in that he urges clarifying and foregrounding the associations users will make when encountering any interface (from a light switch to a Web credit card payment form). But both Norman and Nielsen ultimately suggest this lowest-common-denominator approach that reinforces a Control model of design. While appearing to be "user-centered," Norman recommends anticipation of user expectations. While Nielsen recommends getting user feedback, he also fervently recommends a "keep it simple" design scheme. When asked about the "average user," Nielsen does correct the interviewer with the preferred term, "representative user" (Head 1999, p. 34). This only begins to acknowledge that design is about context. Unfortunately, though, even "representative user" still frames Web interaction as a one-to-one technology. In organizations, Web interaction happens on a continuum of one-to-one, one-to-many, and many-to-many interactions. For this broad spectrum of communication, designers must look beyond the user and acknowledge that community and culture are the ways in which the experiences and expectations of various "representative users" intersect.