Production of Value Outside the Lab: From the Architecture Machine Group to the One Laptop Per Child Project
A lab organized to ensure profitability is also a lab that sells itself as being dedicated to innovation. For decades, the popular and tech presses have positioned The MIT Media Lab as the first and the most well-known of its kind. Wired staff writer Fred Hapgood wrote in 1995 that “[f]or a few years after it officially opened in 1985, the MIT Media Laboratory may have been the most celebrated research institute in the country, at least as measured by inches of newsprint or minutes of air time. Perhaps it still is.”54 Even in 2019, nearly twenty-five years later, CBS still runs stories about how the lab is a “Future Factory” where “Tomorrow’s Technology is Born.”55 For a quarter of a century, the lab has been synonymous with inventing the future because of the central role it has historically played in the fields of wireless networks, field sensing, web browsers, and the Web, as well as the role the lab is currently playing in neurobiology, biologically inspired fabrication, socially adaptive robots, emotive computing, and many other areas. Or, at least, the role the Lab asserts it has played. The MIT Media Lab has also become synonymous with the future largely because of a dogged thirty-year marketing campaign whose success we can measure by the fact that almost any discussion of the technology of tomorrow is a discussion about some project at the Media Lab.56
However, the contradictions of a discourse that espouses humanism through technology as the primary mechanism for generating value outside of a lab are already evident in projects produced by the Architecture Machine Group (AMG), founded by Negroponte in 1967. In a section of Negroponte’s book The Architecture Machine titled “Humanism Through Intelligent Machines,” Negroponte lays out the need for “humanistic” machines that respond to users’ environments, analyze user behavior, and even anticipate possible future problems and solutions—a machine that does not so much “problem solve” as it “problem worries.”57 Negroponte seems to be thinking not only of the potential of such a machine in the form of a computer in the home, but of one specifically aimed at children, for “the computer utility will become a consumer item, and every child should have one”–precisely what the One Laptop Per Child computer aspired to be thirty years later.58 Negroponte also seems to be responding to then-MIT President Howard W. Johnson’s call for members of the university community to consider the social impact of their work to a greater degree. Negroponte asserts that the aim of such responsive architecture machines is “definitely humanization.” He continues: “It is simply untrue that ‘unpleasant as it may be to contemplate, what probably will come to be valued is that which the computer can cope with—that is, only certain kinds of solutions to social problems.’ We will attempt to disprove the pessimism of such comments.”59 In short, computers ought to make everyone more human, whatever that meant, not less (or other), and they ought to be able to anticipate and solve any social ill.
His example of what such an adaptive, responsive machine could look like as it actively attempts to “problem-worry” comes from an experiment that undergraduate Richard Hessdorfer undertook in the lab. Writes Negroponte, “Richard Hessdorfer is . . . constructing a machine conversationalist . . . The machine tries to build a model of the user’s English and through this model build another model, one of his needs and desires. It is a consumer item . . . that might someday be able to talk to citizens via touch-tone picture phone, or interactive cable television.” To help build this machine conversationalist, Hessdorfer decided to bring teletype devices into a neighborhood in the south side of Boston that Negroponte calls “Boston’s ghetto area.” While the project description never mentions race, documentary photographs that appear throughout The Architecture Machine of residents using the machines feature an African-American man.60 The description below implies other users might have been immigrants or simply non-English speakers. Negroponte writes:
Three inhabitants of the neighborhood were asked to converse with this machine about their local environment. Though the conversation was hampered by the necessity of typing English sentences, the chat was smooth enough to reveal two important results. First, the three residents had no qualms or suspicions about talking with a machine in English, about personal desires; they did not type uncalled-for remarks; instead, they immediately entered a discourse about slum landlords, highways, schools, and the like. Second, the three user-inhabitants said things to this machine they would probably not have said to another human, particularly a white planner or politician: to them the machine was not black, was not white, and surely had no prejudices. (The reader should know, as the three users did not, that this experiment was conducted over telephone lines with teletypes, with a human at the other end, not a machine.)61
While the project was supposed to be an example of what Negroponte calls an “adaptable machine” (and thus an architecture machine), it was instead an elaborate sleight-of-hand involving fairly standard telecommunications equipment for the time.62 As Orit Halpern notes, the experiment was at best performing “a future ideal” of interactive media communication.63 The above passage also demonstrates the contradictions—and profound problems—of a humanism via technology that grants agency to (for Negroponte, “humanizes”) certain humans over others (presumably the white, remote observers over the non-white and/or immigrant humans who the narrator assumes will behave poorly and type “uncalled-for remarks”); grants agency over technology by some rather than others; and, moreover, under the guise of a laboratory experiment, does so via a deception that presents the teletype machine to research subjects as somehow responsive to non-Whites’ and/or immigrants’ expressions of concern about living conditions.64 It also clearly demonstrates what can happen when we believe so completely in the neutrality of the machine—or its assumed capacity to give us pure, unmediated access to reality—as to think it can be called on as a magical, mechanical solution to any human problem. As Halpern writes: “Architecture Machine Group attempted to turn the external traumas of American racism and economic crisis into an interactive simulation and to advance computing as a solution to these structural problems.”65
The second AMG project was just as problematic. This time the subjects in the experiment were not African Americans and immigrants, but gerbils. The experiment, referred to as Seek, was part of a 1970 exhibition at the New York Jewish Museum called SOFTWARE. It consisted of a computer-controlled environment full of small blocks and gerbils, all of which was contained by Plexiglass. The gerbils were there to change the position of the blocks and a robotic arm was then supposed to analyze the gerbils’ actions to try to complete the rearrangement according to what the machine thought the gerbils were trying to do. The catalogue for the exhibition describes Seek as follows:
Seek metaphorically goes beyond the real-world situation, where machines cannot respond to the unpredictable nature of people (gerbils). Today’s machines are poor at handling sudden changes in context in an environment. This lack of adaptability is the problem Seek confronts in diminutive. If computers are to be our friends they must understand our metaphors. If they are to be responsive to changing, unpredictable, context-dependent human needs, they will need an artificial intelligence that can cope with complex contingencies in a sophisticated manner . . . much as Seek deals with elementary uncertainties in a simple-minded fashion.66
Not surprisingly, with its simultaneous lack of interest in the needs and particularities of gerbils and its equation of gerbils with “people,” the experiment was a disaster. As Orit Halpern puts it, “The exhibition’s computers rarely functioned . . . the museum almost went bankrupt; and in what might be seen as an omen, the experiment’s gerbils confused the computer, wrought havoc on the blocks, turned on each other in aggression, and wound up sick. No one thought to ask, or could ask, whether gerbils wish to live in a block built micro-world.”67 Such experiments included multiple implicit assumptions about “people” (and other intelligent animals) as part of technological systems but it was also, very much, part of the marketing traditions of technological spectacle.68 Like Claude Shannon’s maze for Theseus the mouse (see the “Imaginary” chapter), it seems to be mostly about showmanship as a way to generate notoriety, funding, and, eventually, profitability. Even Negroponte himself later describes these early projects by the AMG as stunts, rather than as thoughtful interventions, in that they were “using museums and exhibits was a way to do sort of the outrageous ‘cause you had a pass where you could do things and you don’t necessarily have to justify them in some scientific context and we were allowed to play”.69
In the years after Seek and before the opening of the MIT Media Lab in 1985, this pattern of promoting humanism via technology continues in a series of projects by Negroponte’s colleagues and collaborators. The projects were also instrumental in building up to the methodologies of “demo or die”, testing, and experimenting that became integrated as key lab techniques (see also Techniques chapter).70
What these projects and their techniques have in common is that they treat both humans and technologies as abstractions, grant agency to some humans over others, and thereby grant agency over technology to some rather than others. Continuing the push Negroponte described in The Architecture Machine to develop a personal computer that was appealing to children, fellow MIT faculty member, Media Lab affiliate and eventual co-founder of the OLPC project Alan Kay launched the “interim DynaBook” in 1972. Similar to the Hessdorfer experiment’s assumption that teletype users would simply figure out how the machine worked, without any guidance or instruction, while simultaneously disempowering these same users by subjecting them to an elaborate deception at the hands of remote (white) researchers, Kay launched his project at the ACM National Conference in Boston by declaring his belief that children “learn by doing” and that, unlike “the African child,” American children lack meaningful opportunities for learning by doing: “Unlike the African child whose play with bow and arrow INVOLVES him in future adult activity, the American child can either indulge in irrelevant imitation (the child in a nurse’s uniform taking care of a doll) or is forced to participate in activities which will not bear fruit for many years and will leave him alienated.”71 With the vast variability of cultures across the continent of Africa, not to mention the vast variability in hunter-gatherer techniques and methods of teaching children, once again lab discourse utilizes an abstraction—“the African child”—to promote a personal computer. Even stranger and more ironic is the fact that Kay seems to have believed he was going against the grain of MIT President Howard W. Johnson’s exhortation to fix social problems with “humane technology” and Negroponte’s advocation for “humanism through technology.” He writes, “For many years it has been a tradition to attempt to cure our society’s ills through technology. . . Unfortunately, most of these ‘cures’ are no more than paint over rust; the sources of the initial problems still remain.”72
For Kay, the initial problem is the learning process that he assumes is informing most technological cures. Inspired by work by John Dewey, Jean Piaget and Seymour Papert, he asserts that more attention ought to be paid to the child as “a ‘verb’ rather than a ‘noun,’ an actor rather than an object; he is not a scaled-up pigeon or rat” (or a gerbil).73 The DynaBook was designed, from the ground up, to allow this active, empowered child to learn “algorithmic thinking.” With a proposed price of $500 (a tiny fraction of the price of any computer being sold in the early 1970s (but about $3000 at the time of this writing when adjusted for inflation), ought to be “within reach of millions of potential users.” However, whether American or African, the figure of the child is yet another abstraction that existed outside of any specific socio-economic context. While Kay may not have been making assertions about the neutrality of the machine as Negroponte did alongside later Media Lab directors, his reliance on the child as an abstraction points to how he cannot conceive of how bias is inevitably part of the design of any technology, including the DynaBook.
In Negroponte’s second volume on the work of the AMG, The Soft Architecture Machines (1974), the leveraging of abstract, de-contextualized versions of non-white, non-adult, and/or non-western communities to promote particular digital technologies continues. Negroponte draws on Bernard Rudofsky’s well known Architecture without Architects (1964), which “attempts to break down our narrow concepts of the art of building by introducing the unfamiliar world of unpedigreed architecture”(i.e. building practices that are “vernacular, anonymous, spontaneous, indigenous, rural”).74 From this foundation, Negroponte extrapolates the idea of “the indigenous architect as an archetype.”75 This figure, according to Negroponte, lived in an environment that was “simple and comprehensible, punctuated with limited choices and decisions. He no more needed a professional architect than he needed a psychologist or legal counselor.”76 By contrast, he continues, “in our fast-moving societies our personal experiences are phenomenally varied . . . This is why we need to consider a special type of architecture machine, one I will call a design amplifier.”77 According to Negroponte, his 1973 project Urban5 was an ideal example of a design amplifier. Essentially a more sophisticated version of the Hessdorfer Experiment and Seek, it was a computer-aided design program that allowed the user to manipulate virtual cubes with a light-pen while engaging in dialogue with the program (which drew from a dictionary of 500 possible options and questions) about their architectural wants and desires.78 On the surface, Urban5 sounds like a perfectly reasonable, early experiment in machine intelligence. However, once we know it is framed by the figure of “the indigenous architect”—whose fictionality becomes apparent via assertions about how “he” lives in a simple and comprehensible world lacking complexity, not to mention any kind of cultural and geographical specificity—Urban5 looks structurally identical to the Hessdorfer Experiment, Seek, and the DynaBook in creating problematic racialized figures in the context of primarily white elite corporate university practices, techniques, and experiments.
By the late 1970s, the turn toward leveraging people of color, and/or non-adults, and/or non-westerners as a way to promote particular digital technologies shifted to another complex abstraction, namely “The Third World.” Following the massive success of The American Challenge, Servan-Schreiber went on to publish The World Challenge in 1981—the same year he founded the Centre Mondiale Informatique et Ressource Humaine in Paris and appointed Negroponte as its founding Director. This politician-turned-journalist-turned-tech-advocate describes a “renewed fraternal endeavor” that took place in Riyadh, Saudi Arabia. Its participants included oil ministers from middle-eastern countries and unnamed, unspecified “pioneers” who we can only assume are tech entrepreneurs from the west.79 The point of the meeting, called the Taif seminar and chaired by Servan-Schreiber, was to discuss a secret report which was supposed to be made public by Sheikh Yamani on the twentieth anniversary of the founding of OPEC, but never was.80 According to Servan-Schreiber, this report “aims for nothing less than a new alliance between the Arabs and the peoples of the Third World against their traditional exploiters, the industrialized West. It is conceived as a warning, a challenge and finally a demand for a massive transfer of technology from the United States, Europe and Japan to the poor and needy”.81 While this sounds admirable enough, by the end of the book, Servan-Schreiber slips into American-biased rhetoric about the conjunction of “Third World peoples,” telecommunications, microprocessors, and the importance of individual (read “American”) learners “learning how to learn”—a concept presumably gleaned from the same trio of thinkers influencing Alan Kay (Dewey, Piaget, and Papert). In short, “[t]he peoples of Asia, Africa and Latin America should not have to repeat this process [of heavy industrialization]. Telecommunications, microprocessors and their tendency to converge in the new creative process should be placed freely and completely at the disposal of Third World peoples—so that they can become creators themselves.”82
By comparison with more egregious technology projects in the name of humanism, Servan-Schreiber’s account of the Taif seminar and of the technology transfer he believed ought to happen in the name of “the Third World” is not nearly as problematic. However, it is noteworthy for its dependence once again on leveraging an abstraction, “Third World peoples,” to use rhetoric of technological innovation to create new markets for products. It also paves the way for the OLPC project insofar as the first project Negroponte takes up in the early 1980s as founding director of the Centre Mondiale Informatique et Ressource Humaine is the design of a pilot project for “computers-in-education for developing countries.”83 Negroponte and Seymour Papert worked to bring this pilot project to Pakistan, Columbia, and Senegal, where they installed several hundred Apple II computers donated by Steve Jobs. As Negroponte describes it in 2005, “for a time, these school kids commanded more computing power than did the central Senegalese government.”84 While the project only lasted for a year, Negroponte continued to launch similar computer campaigns, including a nationwide program to install computers in Costa Rica’s primary and secondary schools in the mid-1980s, and a program he launched with his son in 1999 to send 50 laptops to schools in a rural village in Cambodia.85 Negroponte’s tendency to rely on Cambodia to illustrate the power of his computer projects also involves frequent resort to the abstractive logic of humanism via technology at work, as his anecdotes always willfully ignore the particularities of individual cultures and communities for the sake of advocating for the mass exportation of technology projects.86
After traveling the world since at least the early 1980s to sell personal computers to developing nations, Negroponte announced in 2005 he had created a non-profit organization, One Laptop Per Child, to produce a $100 laptop “at scale”.87 In other words, the cost of the laptop could only be this low in the early 2000s if, according to Negroponte, they could amass orders for 7–10 million machines. Despite his oft-repeated statement that OLPC was not a laptop project but rather an education project, its essence was the same as the Hessdorfer Experiment and Seek: got a poverty problem? Get a computer!88 A year after the program’s launch, Negroponte continued to rely on Cambodia as an example, touting how the children there learned “Google” and “Skype” as some of their first words while their parents “love the computers because they’re the brightest light source in the house.”89 He goes on to sell the project by declaring that its rightness and goodness is so unquestionable that “[t]his laptop project is not something you have to test. The days of pilot projects are over. When people say well we’d like to do three or four thousand in our country to see how it works, screw you, go to the back of the line and someone else will do it, and then when you figure out this works you can join as well.”90 By 2012, studies were clearly indicating that whether they were used in Peru, Nepal, or Australia, the laptops made no measurable difference in reading and math test scores.
The pitch-perfect ending to this story is that in 2013, after selling millions of laptops to developing nations around the world, laptops that made no measurable improvement to anyone’s lives, Negroponte left OLPC and went on to chair the Global Literacy X Prize as part of the XPRIZE Foundation. However, the prize itself no longer seems to exist, and just a year later (2014), there’s no record of him being with the organization. XPRIZE, however, does exist, and has set new records for the density of humanist sloganeering:
XPRIZE is an innovation engine. A facilitator of exponential change. A catalyst for the benefit of humanity. We believe in the power of competition. That it’s part of our DNA. Of humanity itself. That tapping into that indomitable spirit of competition brings about breakthroughs and solutions that once seemed unimaginable. Impossible. We believe that you get what you incentivize . . . Rather than throw money at a problem, we incentivize the solution and challenge the world to solve it . . . We believe that solutions can come from anyone, anywhere and that some of the greatest minds of our time remain untapped, ready to be engaged by a world that is in desperate need of help. Solutions. Change. And radical breakthroughs for the benefit of humanity. Call us crazy, but we believe.91
The board of XPRIZE board includes every major corporate executive one can think of and it appears they are not even aiming to produce things anymore, but just “incentives.” In many respects, it’s the logical conclusion of the trajectory toward abstraction we have outlined.
At the time of this writing, The Media Lab itself still exists. In spite of the lab’s tarnished image as a result of its affiliation with Jeffrey Epstein, it is busy churning out demos and products for consumers, corporations, and the military. It is also full of people—students, faculty, administrative staff, lab techs and others—who were not privy to the benefits of its entrepreneurial showmanship, nor desired to be. Indeed, also in the accounts of people involved with the Media Lab there has been important commentaries as to how the Media Lab’s modus operandi and rhetoric of technological humanism combined with solutionism has looked like. For example M. R. Sauter of the University of Mariland iSchool, who worked with Ethan Zuckerman at the Media Lab as an MA student, notes how the financial structure was directly linked to how some of the high profile projects were operating: “The Media Lab was founded with the explicit intention of, through funding structures, exempting its researchers from the accountability models inherent in competative [sic] academic and research granting and corporate research contracts. . . .”92 They continue:
That “freedom” to throw ourselves into the future with little to no accountability had a huge cost, not just to students and faculty but also to people we have never met because disproportionate personal and corporate fortunes are always taken out of the backs of *someone*.93
However, this does not imply that the Lab’s projects were all compromised—quite the contrary. 94 Ito’s notes about MIT Media Lab’s Lab Culture as a complex self-adapting system is here an interesting way to perhaps not resolve the question but to point to how the self-description has tried to account for the existence of multiple projects and their very different stakes, ethics, and methods. But the situation does bring to the fore the internal contradictions in many of the larger scale hybrid labs that have to be acknowledged in order to understand how the practices, discourses, infrastructures, and policy contexts of labs are more complex than mission statements and manifestos lead to believe. While “follow-the-money” methodology might be useful in bringing out some of the most problematic sides of current neoliberal university system, we also need to understand that hybrid labs are not reducible to stories about their (often white male) directors, and the narrative of their histories has to account for the complexities and alternative examples. This is also why we want to turn to a different kind of a Media Lab that emerges in the wake of MIT Media Lab, one that engages with an alternative formulation of articulating people, institutions, and lab culture.