Usable and Useful: On the Origins of Transparent Design in Personal Computing

Introduction

With respect to computer interfaces, transparency of interaction serves as a primary goal of design. Despite other meanings of the term, “transparent” design does not lead to software that exposes its inner mechanisms to users. On the contrary, transparent design holds that complex systems must be concealed: a usable interface is one that sublimates technical information about a system’s functions and instead provides users with a simpler, “intuitive” model for understanding its operation. Ideally, users of a transparent interface shouldn’t think of themselves as using a computer at all. Transparent design is now so ubiquitous that most public discussion of computing revolves around the things that we can do with computers rather than how computers work. We trust that interfaces present us with a faithful representation of the systems they provide access to or largely assume that those mechanisms they mask are hidden from us for good reasons. There are compelling practical and political reasons for not requiring people to have an advanced understanding of how computer systems work in order to use them; however, this essay pushes back on the naturalizing rhetoric supporting transparent design (Norman 1988; Shneiderman 1983; Winograd and Flores [1986] 1987). Whereas foundational theories of human–computer interaction claim that transparent design is based on universally shared cognitive principles, tracing the social construction of usability in personal computing reveals that much of the generalized language we use to describe our use practices is derived from rhetorical strategies intended to represent particular features of specific commercial products as necessary for or even desirable to users. As I show in this essay, the idea of transparency as a natural form of interaction entered public awareness as spokespersons and advertisements for companies like IBM and Apple Computer attempted to situate their products in response to popular concerns about computer literacy. While promoting their computers as more usable and useful than their competitors, many commercial developers also leveraged a rhetoric of usability to justify design features that were produced via organizational constraints or that were originally intended in part to exercise control over the media ecosystems that would form around their products.

Scholarship in digital media studies has long demonstrated an awareness of transparent design; however, the field has focused largely on transparency’s role in shaping aesthetic experiences. In her description of the flickering signifier, for example, N. Katherine Hayles (1999) argues that software’s complex networks of significations appear to human observers to be full of “unexpected metamorphoses, attenuations, and dispersions.” Making software comprehensible thus requires that its self-representation be necessarily incomplete (pp. 29-31). Other scholars have taken this assumption a step further by arguing that our experience of media generally is predicated upon a technology’s capacity to efface the mechanisms of its aesthetic production (Bolter and Grusin 2000; Galloway 2012). Digital media studies has thus implicitly naturalized transparent design by suggesting that we can only recognize a technology as a medium by virtue of its capacity to focus our attention on the cultural content it conveys rather than on the specifics of its operation. Yet scholars have also suggested that while software must be transparent to be humanly comprehensible, directly engaging with the mechanisms of software’s self-representation is an important part of exploring the expressive affordances of computational media (Manovich 2001; Hayles 2010). In fact, manipulating user perception of software systems by strategically misrepresenting their internal operations is often key to producing compelling cultural experiences through software, particularly in video games (Wardrip-Fruin 2009; Black 2013; Natale 2019). While this work is important in helping us to understand the complexity of software’s strategies of self-representation, digital media studies has largely left open the question of how the commonplace, seemingly unremarkable elements of transparent interfaces influence our personal and collective relationships to computing.

The transparent interfaces that we use every day and the language of usability they support structure how we understand, discuss, and exercise our digital agency. Transparent interfaces naturalize the self-effacement of technological systems, effectively bifurcating computational culture. User access to computational culture occurs through the intentionally limited interactive and discursive frameworks provided to us by designers. Because so many of our common computing practices are predicated upon invisible, automated systems, we largely trust and accept the representations of those systems provided to us via interfaces as a condition of use. Much like scientists use a rhetoric objectivity to sublimate discussion of the cultural and social influences on their work (Latour 1987), software designers leverage a rhetoric of usability to present their technologies as “gray media” that appear to be “transparent to human intentions” (Fuller and Goffey 2012, 5). We assume that gray media are neutral: invisibly supporting but not influencing our digital agency. However, there is no guarantee that the seemingly self-evident system representations provided to us match the algorithmic operations occurring beneath the interface. Gray media have not been a focus of humanities-based approaches to digital media studies because they seem to lack a “hidden meaning that might be brought to light (they need not be the object of a hermeneutics)” (Fuller and Goffey 2012, 12). Yet, it is important to recognize, as Wendy Chun has argued, that software’s focus on accepting surface-level representations makes them “functionally equivalent” to ideology. Latent within the promise of action and agency that software interfaces present to us is an “authoritarian sanction—one is given the right to think X, one may think X, in which case the authority would be the machines themselves, our supposedly loyal servants” (Chun 2011, 79). In this sense, we need to recognize that transparent design’s focus on the things that computers can help us to do masks the way that computers structure those activities in ways that benefit their designers. Although hacking is possible, most of us use our computers as their designers intend for us to. If the goal of usability is to align user behavior with an intended set of acceptable practices, then we need to recognize that its goal is also to minimize our repurposing of technology. Tracing the social construction of transparent design can thus help us to better understand a host of assumptions latent in the rhetoric of usability about the role computers should play in our culture, who should have access to information about computer systems, and the political power of design.

My focus in the pages that follow is on two machines that introduced two influential models of usability in the early to mid-1980s: the IBM 5150 “Personal Computer” (or PC) and the Apple Macintosh. Each machine presented users with a different form of interaction, yet both companies framed their products as responses to popular anxieties about computer literacy. After first highlighting a disconnect between popular memory and mainstream press coverage of the “personal computer revolution” during the mid to late 1970s, I examine how publications written toward a variety of audiences raised concerns that computer literacy requirements could be a potential threat to the long-term economic viability of personal computing. In this respect, IBM and Apple each attempted to portray its products as responsive to concerns that the average American would be unable to learn how to use a computer well enough to justify buying one. Both companies incorporated elements of the popular discussion of computer literacy into their promotion of their computers. But whereas IBM promised that the PC would make it easy to develop computer literacy, Apple claimed that the Macintosh’s usability model effectively made computer literacy obsolete. By showing how each company crafted its representations of usability to account for internal and external influences on the technical development of their computers, I argue that rhetorics of usability can serve to reframe design decisions resulting from organizational constraints and economic strategies so that they appear to have been made in the best interest of users.

Public Perceptions of Computer Literacy in the Early 1980s

During the mid-1970s, just prior to the availability of the first personal computer kits, many programming enthusiasts argued that computer literacy was key to resisting the ability of technocrats to leverage computer systems as sources of political power. Ted Nelson’s writings, in particular, suggested that computing’s cultural potential had been held back by corporate bureaucracies. Computing could support both existing and not yet realized forms of cultural and social activity when placed into the hands of ordinary people. The two keys to the coming “personal computer revolution,” he suggested, were making knowledge about computers as freely available as possible and giving people the opportunity to experiment with that knowledge free from the constraints imposed by the Kafkaesque bureaucracies of “computer priesthoods” (Nelson 1974). Nelson and others, in short, argued that changing the way we access computing power would effectively humanize computing by opening it up to interests other than those that narrowly benefited technocratic institutions. West Coast groups like the People’s Computer Company and later the Homebrew Computer Club are often the focus of discussions of early personal computing, but other, less well-remembered groups formed across the country. Regardless of their structure or the specifics of their mission, hobbyist communities and users groups during the 1970s generally demonstrated a commitment to cultivating the computer literacy of their members, offered assistance or advice with the difficult process of assembling a working computer from a kit, and compiled information about local computing resources.

Although the influence of hobbyist culture looms large in our popular memory of the personal computer revolution, it was not discussed much at all in the mainstream press until the late 1970s. Instead, most discussion of the potential of microcomputers during that decade presented a much more conservative vision of revolution, emphasizing the ways that desktop-sized computers would automate our offices and homes (Haigh 2006). Prior to the rise of retail-friendly personal computers, journalists writing in this framework typically referred to executives from mainframe manufacturers like IBM or its chief competitors—Burroughs, UNIVAC, NCR, Control Data Corporation, and Honeywell (collectively referred to in the press as BUNCH)—to describe the revolutionary potential of new, desktop-sized computers. An article published in the Washington Post, for example, includes quotes from executives representing several of these companies explaining how they would soon offer smaller computers that were “friendly” and “personal” enough to be used for a variety of purposes in offices, factories, and homes (O’Toole 1977). Following the first announcements of “appliance” computers in late 1977 by companies like Apple, Tandy, and Commodore, the press began to rely less on executives from IBM and BUNCH to describe the coming personal computing revolution. Nonetheless, the popular focus on business and domestic friendly uses remained. Corporate histories of Apple, Tandy, and Commodore suggest that, at least internally, executives wanted their products to be seen as alternatives to hobbyist computing (Bagnall 2012; Farman 1992; Moritz 2009). Unlike the kits that had been the focal point of mid-1970’s hobbyist culture, each of these companies promised that their machines would not only be usable for a variety of tasks at home, school, or the office but also useful right out of the box.

Despite manufacturers’ promises that preassembled computers could be treated like appliances, many general-interest news magazines published accounts of user experiences suggesting that novices continued to find personal computing to be overwhelming. In retelling his experience with a Commodore PET and an Apple II, for example, Peter Martin writes in Money that he was ultimately unable to do much more than run the demonstration software included with his machines on cassette tape and copy some simple BASIC programs out of a magazine. Getting the computer to live up to the life-changing promises he had read about was “beyond [his] grasp and beyond the grasp of anyone else who doesn’t intend to take up computer programming as a serious hobby” (Martin 1978, 86). Stories like Martin’s can be found well into the early 1980s. In February 1982, for example, Newsweek’s cover story, “Home Is the Where the Computer Is,” was a mixture of reporting on new personal computers available to consumers and futurist descriptions of what would soon be possible with machines like them. Nested within it was a two-column story by Elizabeth Peer describing the moments of triumph and disappointment she experienced working with some sample machines sent to their office. She concludes that “after a total of 70 hours with the computers,” she felt that she had learned quite a bit about how they worked but still “remained wholly incapable of utilizing them to suit the needs [she] would have bought them for” (Peer 1982, 53). Although the popular press frequently proclaimed during the late 1970s and early 1980s that the personal computer revolution had now, finally, arrived, it was not uncommon to find moments of doubt or concerns about whether or not the average American could comfortably incorporate them into their home and working lives.

Some early observers of the growth of the personal computer industry even went so far as to assert that the United States may soon face a “computer literacy crisis.” One of the earliest mentions of a computer literacy crisis can be found in a National Science Foundation (NSF) study of science education outcomes. Under the assumption that computers would soon be integrated into almost every aspect of professional life, the study concludes that there is “national need to foster computer literacy” lest American science and industry fall behind the rest of the world, not to mention the “unacceptable social and psychological costs” of those individuals whose illiteracy leaves them unable to compete for jobs in a changing economy (Molinar 1979, 280-83). A psychological study on the influence of computers on white-collar organizational culture in the early 1980s similarly identified the perceived need to relearn how to perform familiar tasks using a computer as a significant source of anxiety that affected everyone from clerical workers to office managers to executives (Brod 1984). Some columnists in computing magazines expressed similar concerns, cautioning their readers to recognize that others may not share their own enthusiastic embrace of and comfort with computers. Writing for InfoWorld, Ed Martino declared that a “crisis of epidemic proportions in computer literacy exists in this country.” Even though there seemed to be a consensus that everyone should integrate computers in their offices as soon as possible, “[v]ery few business and professional people know enough about computers to cope with the emerging ‘information age’” (Martino 1980, 10). In diagnosing the cause of this potential crisis, Martino argued that too much of computer education was focused on highly technical practices like programming. Instead, he concluded, computer literacy training should focus on specific tasks and software applications, so that people would learn to see computers as usable and useful rather than as something to be mastered or explored.

The focus of computer education on programming also helped to create the popular impression that learning to use a computer required a great amount of time and dedication for minimal reward. Columnists like David D. Thornburg suggested that “computers [were] being sold to many thousands of people who have no idea what they are getting into.” Despite the explicit promise of immediate benefits and recognizable purposes, most people “generally [had] little understanding of the effort needed to make computers do truly useful things” and “could save a lot of grief by flushing the money [they would use to buy one] down the drain instead” (Thornburg 1981, 11). Writing for the Technology Review, Robert Cowen similarly suggested that the press had for the past few years been talking as if the personal computing revolution had come to fruition when in practice almost nothing had changed. The software available to consumers felt “tedious” and the tasks they assisted with were “trivial.” Computer hardware manufacturers and software developers had embraced the idea of revolution as a sales tactic but had in practice performed little more than lip service with respect to helping the general public realize tangible benefits from computing. Actually achieving the kind of broad cultural and social benefits of personal computing described in the popular press would require a “deep understanding of how people relate to computers and master them reflected in services, software, equipment, and instruction manuals” (Cowen 1981, 20). Extending his discussion into a later column, Cowen (1982) further argued that “widespread literacy [had been necessary] for the printing press to make its impact, to say nothing of the growth of the publishing industry. The home computer awaits a similar rise in computer literacy and an information industry that caters to it. These will be the real agents of change” (p. 7). Although it is difficult to say how well concerns like these reflected attitudes of individual consumer, they nevertheless appeared across a wide variety of publications. The plethora of advertisements during the late 1970s and early 1980s claiming that new products were “user-friendly” and “easy to learn” suggests that hardware and software designers were prioritizing usability; however, as I will show below, the rhetoric of usability they leveraged to promote their products was often in practice a strategy to reframe design features resulting from organizational constraints and economic strategies as responses to popular concerns about computer literacy.

The IBM Personal Computer

Compared to the Altair 8800 and the Apple II, the IBM PC has received comparatively little attention in books discussing the first decade of personal computing. When it is discussed, accounts typically focus on IBM’s “surprising” decision to deviate from the business strategies focused around proprietary technologies that had defined the company for decades (Freiberger and Swaine 2000, 328-37). Yet accounts from engineers who contributed to the PC’s design suggest that the machine’s open architecture was more a product of a tight time line than an ideological change within the company. When directed both to develop a prototype and to prepare it for retail within eighteen months, the PC’s design team members concluded that they would not have time to design proprietary hardware or software (Chposky and Leonsis 1988; Porter 1984; Bradley 1990). Many reviews for the PC thus found its hardware and software to be unremarkable. A review appearing in BYTE magazine, for example, described the machine as synthesis of best practices: a computer that does “everything conventionally but well” (Williams 1982, 65). Yet BYTE’s review also states that the included documentation “will set the standard” for the industry: “Not only are [the included manuals] well packaged, well organized, and easy to understand, but they are also complete” (Williams 1982, 56). When considered alongside the popular concerns about usability outlined above, IBM’s documentation becomes a strategy for making an unexceptional user interface stand out as somehow more usable and useful than that of its competitors. Unlike other computers, IBM promised that the PC was specifically designed to help users develop the computer literacy skills necessary for them to be comfortable with computing.

Praise for the PC’s documentation does not mean that other manufacturers were not supplying their users with detailed reference manuals. In 1978, Apple began including with its Apple II a volume titled the Apple II Reference Manual. While the manual did include some basic setup instructions and introductory programming lessons for novices, the depth of detail in the majority of its chapters suggests that it was written for hobbyists who already possessed a sophisticated computer literacy. In 1979, Apple began swapping the Apple II Reference Manual for The Applesoft Tutorial, a book presented as more appropriate for novice users. Yet, as an article comparing Apple’s and IBM’s documentation notes, The Applesoft Tutorial “goes overboard…. cramming its pages with hardware and software details that could clog the circuits of any beginner’s brain” (Hewes 1982, 116). Despite including illustrations, colored fonts, and more casual prose than the Apple II Reference Manual, The Applesoft Tutorial’s pages were extremely text heavy. Its lengthy descriptions were likely intended to model a more conversational style; however, the writing often takes a condescending tone. For example, the manual at various points refers to technical terms as the language of “computerniks,” reminding novice readers that they are not a part of computing culture (Apple Computer Inc. 1978, 14, 17).

By comparison, IBM’s documentation for the PC is split into multiple volumes that are organized around a progression from novice to advanced use. When the PC was first sold to the public, its standard package included a three volume “Personal Computer Hardware Reference Library”: the Guide to Operations that served as a general reference and setup guide, the BASIC manual that offered instruction in programming concepts alongside a complete language reference, and the Disk Operating System manual that explained file system operations. Additionally, users could purchase separately the Technical Reference that provided detailed information about almost every aspect of the machine’s hardware. While they are not formally numbered, the Guide to Operations manual effectively serves as volume one in the series, and other manuals direct readers to familiarize themselves with it before consulting the more detailed information they provide. As its introduction notes, the Guide to Operations is written to serve as an introduction for readers who are “new to computers” and “is written in an easy-to-understand language” (p. iv). Compared to Apple’s documentation, the Guide to Operations is much more concise in its descriptions. Instead of lengthy explanations, the Guide to Operations makes heavy use of illustrations. While many of these illustrations are similar to Apple’s in that they document action sequences for routine tasks, they also serve to visualize general computing concepts such as how floppy disks store data or the various outcomes of a COPY command (see Figure 1). Although there is some overlap between the three books, reviewers noted that the introduction to BASIC and DOS found in the Guide to Operations is framed in a way that “assures that a novice can take advantage of the disk operating features and write simple programs using only the elementary manual” (Hewes 1982, 116). The introduction to its chapter on BASIC, for instance, explicitly notes that its goal is to “familiarize [readers] with enough BASIC so that [they] can use the SAMPLES program.…and can run other BASIC programs that [they] can buy.” Thus, readers were given a sense that they could learn enough to make use of their machines without necessarily having to learn to write their own software. IBM thus promoted the usability of its machine by emphasizing that even a very basic level of computer literacy was enough for most people to find the PC usable and useful.

OPEN IN VIEWER

Figure 1.

A page from IBM’s Guide to Operations manual illustrating basic computational concepts for novice users. Source: International Business Machines (1981).

IBM’s early advertisements of the PC similarly emphasized that the machine was designed to ease popular concerns about usability. In its first ad campaign for the PC, “The IBM Personal Computer and Me,” the company announced that their new machines were being built “just for you,” so that you would “have a personal interest in” them. ¹ These advertisements tried to situate the PC in the home, showing individual members of a white, middle-class family with text suggesting that each was not only able to learn how to use a PC but also capable of finding their own uses for one. Although the images and text varied, each similarly suggested that the computer and its documentation were written “in your language, not in ‘computerese.’ Our software involves you, the system interacts with you as if it was made to—and it was.” ² Advertising that promised software would be less confusing than a competitor’s was nothing new, but what is important to note is that IBM tried to represent the PC as accommodating the distinct needs of its users rather than simply making it easier for users to adapt themselves to computing. IBM promised that its documentation offered a “structured learning process” that would get users started quickly. Rather than wading through the turgid prose of reference manuals for a few months, “you can be running programs in just one day. Maybe even writing your own programs in a matter of weeks.…you’ll discover more than the answers and solutions you seek: you’ll discover that getting there is half the fun.” ³ In short, IBM’s advertisements for the PC sought to diminish any anxiety consumers might have about personal computing by creating the perception that the PC’s designers were prioritizing the needs of novices over hobbyists and other more advanced users.

Comments from executives echo many of these ideas; however, their discussions of specific design decisions also show how the company tried to reframe its economic strategy of supporting third-party development using the inclusive model of usability promoted by Nelson and others during the late 1970s. In a 1983 interview with BYTE, IBM’s president of the “Entry Systems Division,” Philip Estridge, continually suggested the company’s goal with the PC was to “provide a simple system that offered customers the ability to experiment with very little effort” (Curran and Shuford 1983, 88). He also stressed that anyone should be able to benefit from the use of a personal computer, regardless of their technical background or prior experience with computing. When asked about who had been the target user during the design process, Estridge explained that IBM doesn’t “have a typical user [in mind] because the machine is so communal that typical doesn’t have meaning, except for the fact that more and more people are discovering that they have needs that can be answered rather nicely with a personal computer. And they are in all walks of life—all the way from very young children to very elderly people—in every profession” (Curran and Shuford 1983, 89). Estridge’s comments here are remarkably similar to the kind of inclusive vision of personal computing that politically minded hobbyist revolutionaries like Nelson promoted during the 1970s. Yet, when considered in the broader context of the interview, IBM’s presentation of the PC and its documentation as an attempt to make computing more accessible to everyone is actually part of a conservative business strategy. Estridge admits, for example, that his team had assumed “personal computer usage would grow far beyond any bounds anybody could see” (Curran and Shuford 1983, 88). Whereas appliance computers of the late 1970s had been promoted using a specific set of uses that journalists like Cowen found unimpressive given the high barrier of computer literacy required to realize them, the PC’s emphasis on developing a working literacy implied that novices would come to see the machine as more useful over time as they learned enough to comfortably use software written by more advanced users. Estridge (1984) returns to this point in a later essay published in Creative Computing, explaining that “it is the element of choice that personal computing offers to people that ensures the continuing appeal and acceptance of personal computers” (p. 194). He thus attributes the PC’s success to the “thousands of applications available”—and made possible by the PC’s open design—that allow users to do things now “that were not possible without a PC” (p. 197). The more comfortable users were with their PC, the more likely they were to continue buying hardware and software for it. And while the PC’s open design and thorough documentation meant that third parties could provide those add-ons, IBM had also developed its own line of first-party peripherals and signed deals with software developers to publish them under the company’s in-house label.

Although it is difficult to determine how IBM’s messaging affected popular perceptions about usability, the PC exerted a strong influence over the personal computing market in the years that followed. One unintended consequence of using commonly available components was that smaller companies were able to produce PC-compatible clones legally. Although it is difficult to determine just how many PCs and clones were in circulation, journalists have estimated that by 1984 as much as 85 percent of personal computer software available for sale had been developed for use on PC-standard hardware. The rapid expansion of the personal computing hardware and software markets likely had less to do with any specific feature of the PC itself than with the confidence that corporate purchasers had in the IBM brand. Novice users who grew comfortable using a computer at work were thus more likely to purchase one for use at home, potentially a cheaper compatible clone (Rubin and Strehlo 1984). Thus, the success of the PC as a standard ultimately leads to the decline of the PC as a product as IBM lost market share to clone manufacturers.

The Apple Macintosh

As noted above, popular conversations about usability continued to highlight concerns about computer literacy well into the 1980s. Following the PC’s release in 1981, many positioned it as representative of an old-fashioned model of usability and suggested that lingering anxieties about computer literacy could only be addressed by moving away from its design. A 1983 editorial for Personal Computing, for example, gestured toward transparent design by arguing that advancements in graphic user interface design by Apple, VisiCorp, and Xerox were “mov[ing] us towards a future in which computing will be easier than driving an automobile.…And in the new generation of personal computing, you won’t have to know much about how computing works in order to compute” (Kellam 1983, 5). Yet it is important not to conflate graphic user interfaces with transparent design. In an article published in BYTE, Sam Edwards (1983) of Personal Software Corporation, a company whose software was often praised for its ease of use, explained that the next generation of user-friendly software must “offer [users] less. The thinner the manual, the less you’ll have to read to learn the program. The fewer the choices on the menus and prompt lines, the less you’ll have to think about before making your choices. The less a program does, the fewer things can go wrong with it” (p. 132). From this perspective, anxieties about usability could be addressed by narrowing the scope of computing. Rather than inspire creativity, Edwards suggested that limitless possibilities emphasized by manufacturers like IBM paralyzed potential new computer users. Positioning the IBM PC as its foil, Apple would present its Macintosh as an alternative that promised a transparency of interaction by offering a simplified model of interaction focused around a handful of uses explicitly marked as cultural or creative.

Since releasing the Apple II in 1977, Apple had presented itself as the company responsible for realizing the personal computer revolution. As IBM laid claim to the “personal computer” label, Apple moved to distinguish itself from its rival. Apple’s two best known public responses to IBM occurred in 1981 and 1984: the full page “Welcome, IBM. Seriously” print advertisement and the Orwell-inspired television announcement of the Macintosh, respectively. In both, Apple wanted to remind the public that the goal of personal computing was to “put real computer power into the hands of the individual” with the goal of improving “the way people work, think, learn, communicate, and spend their leisure hours.” ⁴ Both advertisements paint IBM as more interested in profits than in fostering creativity. More subtly, Apple continued to include the words “Personal Computer” beneath its logo in its print advertisements, and for a brief period in 1983, revised it to “The Most Personal Computer,” suggesting that IBM’s machines were “personal” in name only (see Figure 2). Yet, despite similarities in presentation, there were significant differences in the way the machines structured their relationships with users. While the Apple II was also considered an open architecture machine, the company wielded a mixture of patents and copyrights to ensure that key portions of the Apple II’s hardware remained proprietary. ⁵ As a result, consumers could not build their own fully compatible Apple II computer from parts or purchase one made by third parties. This distinction highlights not just Apple’s desire to maintain control over its technology but also points to its different approach to usability generally. By the mid-1980s, Apple would begin to frame usable computing as a curated experience defined by designers who truly understood the needs of their users rather than something that users should be able to define for themselves.

OPEN IN VIEWER

Figure 2.

A comparison of Apple’s taglines from 1981 (above) and 1983 (below).

Despite many attributing the Apple II’s success to its comparative openness, Steve Jobs frequently criticized it alongside the IBM PC when describing the need for a black box approach to design. According to one of Jobs’ biographers, he had grown frustrated that third parties had more control over how consumers viewed the usability and usefulness of the Apple II: “We don’t have control, and look at all these crazy things people are trying to do to it. That’s a mistake I’ll never make again” (Isaacson 2011, 138). Jobs was also reportedly skeptical of the idea that novice computer users had any real understanding of what it meant to discover their own uses for their machines, remarking that “customers don’t know what they want until we’ve shown them” (Isaacson 2011, 143). Explicitly contrasting the Macintosh’s usability model with that of the PC and the Apple II in a 1984 interview, Jobs explains that the focus of the Macintosh’s design was pushing computing toward a “point where the operating system is totally transparent. When you use a Lisa or a Macintosh, there is no such thing as an operating system. You never interact with it; you don’t know about it.” Users, he explained, were “much more concerned about what the computer will do” rather than how it works, which was the “right way of thinking about products” (Personal Computing 1984, 243). The Macintosh would define usability and usefulness not as a range of potential benefits but as an immediacy of specific benefits: “a person who is going to be using a personal computer half an hour a day, or maybe an hour a day[;] you will not get that person to spend 40 to 100 hours learning how to use a computer” (Personal Computing 1984, 242). Jobs, in other words, suggested that the PC’s usability model effectively left personal computers disconnected from our everyday lives. Within this framework, having to learn how a computer worked in order to use one always meant that computer use practices would be distinct from the cultural activities with which they were supposed to be integrated.

Apple’s advertisements for the Macintosh similarly framed computer literacy as something that competed with or suppressed other cultural skills and knowledges by linking it to IBM’s past technocratic reputation. Apple first announced the Macintosh with a nineteen-page brochure included in several magazines in December 1983. The brochure proudly announces that ordinary people will no longer have to worry about how computers work: the Macintosh was said to be a computer “for the rest of us.” ⁶ The brochure further suggests that the personal computing revolution had been held back by companies like IBM who insisted that people learn to use computers before benefiting from them: “In the olden days, before 1984, not very many people used computers, for a very good reason. Not very many people knew how. And not very many people want to learn.” Apple quite explicitly frames the PC as a machine that exists in a technocratic domain outside of culture by juxtaposing images of the PC’s monochromatic command line with the multifaceted typography and bitmapped windows of Apple’s graphical user interface. Yet Apple’s most pointed critiques of IBM occur without mentioning its rival’s name: “since computers are so smart, wouldn’t it make more sense to teach computers about people, instead of teaching people about computers?” Despite IBM’s attempts to represent the PC’s literacy-focused usability model as a way for users to define their own relationship to computing, the Macintosh advertisements suggest in text and image that the PC and its clones still force us to engage with computing on someone else’s terms. Apple’s advertisements thus argued that computers would only be usable and useful if they reflected the cultural and social values of the contexts in which they were to be used. Requiring users to develop a specialized literacy meant reshaping those contexts around the usability models demanded of technocratic elites.

While most discussion of the Macintosh’s historical significance focuses on its commercialization of graphic user interfaces, it is important to recognize that its interface was not the sole focus of its reviewers. Many, for example, were concerned that the Macintosh appeared to violate what had become key norms in personal computer design. Almost all reviewers noted, for example, that the Macintosh was limited to a single disk drive. In a time before hard drives, two were considered ideal: one for loading applications and another for data storage. One reviewer explained that the Macintosh’s black box design was confusing, given that Apple had managed to survive when competing against Tandy and Commodore through promotion of the Apple II’s comparatively open architecture: “By precluding easy hardware expansion on the Mac, Apple writes off a major component of its early success: expansion flexibility” (Anderson 1984, 22). Market wisdom suggested that the long-term usability and usefulness of personal computers generally required openness, so that third parties could continually work to invent new uses for them: “Sure, it might take some imagination at first to envision the kinds of [internal expansion] cards the Mac might need. But if an expansion bus were available, people would start to invent them” (Anderson 1984, 22). Similarly, the Macintosh also did not include programming tools. BYTE’s review noted with concern that “no one has sold a computer without [built-in support for] BASIC (or some other language) in years” (Williams 1984, 52). With respect to both hardware and software, then, users would be dependent almost wholly on Apple’s ability to provide its own add-ons or to partner with companies who could. Users could not even service the machine themselves or seek help from the same local serviceman they might have sought out for help with their PC. Even if expansion slots had been available, the Macintosh required a special set of tools, to which only licensed Apple technicians had access. Some reviewers, however, recognized that these decisions reflected discussions about the benefits of narrowing the scope of personal computing: “Apple’s challenge [is] for us to accept these differences in the interest of improved computing ability.” The Macintosh’s closed design violates many of the usability principles we assume computers need to embody to be successful, but we must “accept the challenge and step boldly forward to where Apple asserts computing is heading” (Rubin 1984, 58-59). Even as many reviewers noted that aspects of the Macintosh’s design were anti-consumer—or at least benefited Apple as much as if not more than users—almost all were to some degree willing to overlook them because they were persuaded by Apple’s new vision of usability. While almost every review found at least some aspect of the Macintosh’s transparent design controversial, all agreed that this narrowing of user agency was an important step toward invisibly integrating personal computing into American culture.

Conclusion

When reading design theory and discussions of usability principles, we are often left with the impression that theory informs design. But as I have shown in this essay, early principles of usability developed alongside or following the design process in response to organizational constraints, perceived market threats, and the need, ultimately, to make the results of the process seem responsive to user needs. Even as foundational researchers in human–computer interaction spend hundreds of pages laying out the philosophical and scientific bases for transparency, they inevitably refer to commercial products and even at times the Macintosh itself as a source of inspiration (Shneiderman 1983; Winograd and Flores [1986] 1987, 165). It is therefore critical that we recognize just how much of the everyday language we use to discuss our computing practices and our relationship to computational culture generally is crafted to naturalize commercial software development strategies. Discussions of usability are always political because they are always on some level about trying to encourage users to think and behave in ways that reinforce the perception of a particular product’s utility. The “authoritarian sanction” that Chun and others see in software is subtle but can be felt whenever we switch to a new usability framework. The discomfort that people express when switching between Apple’s OSX and Microsoft’s Windows interface, for example, is not because one is objectively more usable and useful than the other but because we internalize the simplified models for understanding computation that their transparent interfaces provide. Learning to use a new interface for the same tasks means challenging our internalized models by recognizing how designers leverage interfaces to make the visual arrangement of their systems seem necessary or even desirable.

In this sense, there is no computational culture apart from that which exists as the result of our contact with particular hardware and software technologies. Just as scholars in electronic literature and video games studies have argued for the importance of original hardware to understanding the cultural significance of interactive media, we also must recognize that any objections or concerns we have about computational culture or the effect that computers are having on other cultural and social domains originate in specific hardware and software technologies. As gray media, they may not be aesthetically novel enough to be considered “objects of a hermeneutics”; however, there are nonetheless plenty of important assumptions hidden within them about the politics of computational culture. These mundane, gray media technologies must be seen as complex objects of study if we are to reject the simplistic, self-serving rhetorics created by their designers to promote them as consumer goods.

If we want to push back on the authority that designers wield over computational culture, in other words, we need to focus on unpacking how larger problematic trends originate in specific commercial technologies. Some journalists have recently begun to ask whether technology companies have grown so influential over media ecosystems as to act like government institutions. But thinking back to the examples above, we should recognize that they have been trying to do so for a long time. Apple can say that they are making computers easier to use by not allowing users to access internal hardware or to program their own software, but these decisions also gave Apple the ability to exert some control over the media ecology that formed around the Macintosh. Notably, Apple was not the only company to try to control third-party development during the first decade of personal computing. Texas Instruments went so far as to threaten to sue programmers who didn’t agree to a 90/10 fee-sharing agreement if they tried to publish software for its TI-99/4A computer independently, and the company was reportedly developing a system of software authorization checks for future models before the company pulled out of the market in the mid-1980s (Nocera 1984). Unlike Texas Instruments, however, Apple successfully reframed design features that some reviewers considered anti-consumer to actually be in the best interest of consumers. Apple has extended this logic into the present by centralizing software distribution on its device through its App Store, and the company’s threats to remove applications from the App Store have moved other, established tech companies to design their software according to Apple’s demands. ⁷

Recent research in digital media studies has begun, rightly, to call attention to the implications of large, invisible systems. Yet, just as foundational works of scholarship were drawn primarily to electronic literature and other explicitly aesthetic forms of digital media, recent critics of invisible systems are drawn to those that are framed as overtly social (Pasquale 2015; Noble 2018; Vaidhyanathan 2018). They raise important concerns about how developers of these invisible systems challenge traditional institutions and/or actively misrepresent the operations of their platforms. Yet we must apply their same level of skepticism and scrutiny to digital media that seems so mundane as to be culturally and socially inconsequential. Even something as innocuous as a weather reporting app could be functioning behind the veil of transparency as a tool to harvest and sell users’ private data. ⁸ Transparent design encourages us to view so much of personal computing as none of our concern, and yet transparent software structures so much of our daily lives. In naturalizing transparent design, we also begin to naturalize the idea that software designers have the authority to restructure every aspect of our cultural and social behaviors and identities: any practice or behavior that might be invisibly integrated with computers comes under their invisible control. We very well could have advanced into the kind of computational future that early, politically minded advocates for personal computing described. We could have graphic user interfaces that encourage experimentation and modification of technical systems or that support the development of common frameworks for exploring the cultural and social affordances of technology. But because we now rely on transparent hardware and software to conceal and manage computation on our behalf, our highly visual interfaces reveal to us only what designers feel is our best interest—and, increasingly, only in their best interest.