OK Computer

How digital graphics remade the material world

A Silicon Graphics, Inc. workstation featured in Steven Spielberg’s Jurassic Park, 1993. The interface seen here is the 3-D File System Navigator for the IRIX operating system, which, as its name implies, is based on UNIX.


JURASSIC PARK, the 1993 blockbuster sensation, contains a sly, almost Velázquezian instance of mise en abyme. In the eponymous theme park’s futuristic genomics lab and control room, scientists fabricate the ultimate prehistoric spectacle using desktop computers, software applications, and file systems manufactured by the real-life Silicon Graphics, Inc. (SGI). These prominently featured SGI workstations don’t just play an essential role in the film’s narrative structure—allowing for plot progression via on-screen graphic visualization—they in fact reveal the very machines used to produce the film’s digital special effects, including many of its hulking dinosaurs. The diegetic embeddedness of this equipment displays an ironic fidelity to reality: The characters in the movie parody the technological processes by which the film itself is made mimetically plausible to viewers.

This depiction of SGI workstations on-set satirizes one of the cardinal rules of computer-generated images: that they must be realistic enough to erase their own medium-specific technical origins. Relatively successful in adhering to this rule, SGI computers were used in the production of every Academy Award nominee for best visual effects between 1995 and 2002. Jurassic Park’s use of the company’s technology, mimicked by a number of big budget movies of that time, helped inaugurate the rise of digital graphics in Hollywood. That rise occurred in tandem with similar developments across the cultural vernacular, with graphical imagery rearing its head in television, video games, architectural design, and software. The large-scale adoption of computer graphics throughout the culture industries, and their enthusiastic reception by audiences, critics, and consumers, marks a significant shift in popular visual aesthetics in the late twentieth century.

Cover of Jacob Gaboury’s Image Objects: An Archaeology of Computer Graphics (MIT Press, 2021).

But the material and intellectual infrastructure needed to bring about this shift required decades of work. Not simply a technical revolution, it was a conceptual and, perhaps, a phenomenological one as well. Jacob Gaboury locates the genesis of this revolution in the fertile period of research conducted at the US Department of Defense–funded College of Engineering at the University of Utah between 1965 and 1980. His new book, Image Objects: An Archaeology of Computer Graphics, maps the legacy of the Utah faculty and its graduates, as well as their methods, as they migrated from academic research to commercial ventures, which included the founding of SGI in addition to Pixar, Adobe, Atari, Netscape, WordPerfect, and a host of other ventures.

Gaboury’s book paves a way for a discourse on computer graphics independent from the already robust literature on midcentury correspondences between art and engineering, such as those of the Experiments in Art and Technology (EAT) group and the endeavors of Bell Labs, the ICA London’s watershed 1968 “Cybernetic Serendipity” show, the information aesthetics of Max Bense and the Stuttgart school, the systems-theoretic criticism of Jack Burnham, or the ecology oriented design advocated by György Kepes, Charles Eames, and the New Bauhaus. Whereas the latter largely circumscribes computer-generated images to the lineage of artistic modernism, Image Objects situates the field’s idiosyncratic strategies within the realm of popular culture and everyday experience. Gaboury thus seeks to recuperate the history of computer graphics from the dominant art-historical and technological disciplines, such as those of film and photography, that tend to subsume it, and in doing so makes a series of bold claims, chief among them being that graphics played a pivotal role in the reorientation of the computer from a logical and mathematical tool to the graphical and interactive medium it is today. For that reason, the book is a unique interrogation of the contemporary optical regime, structured as it is by black boxes and screens.

Facade of the Silicon Graphics, Inc., headquarters in Mountain View, California, 1994 Photo: Paul Warchol.

Gaboury’s book can be considered an “archaeology” for two reasons. First, Image Objects ends, as the author notes, where most histories of computer graphics begin—with the advent of the first mass-manufactured Graphics Processing Unit (GPU) in the early 1980s—and thus constitutes a kind of prehistory. Second, each of its five chapters excavates a particular technical object, most of which form the conceptual basis of tools still used today. These objects are not always material in the strict sense of the term: One chapter examines an algorithm called the “Z-buffer,” employed for representing depth values, while another recounts the emergence of a computer programming model partially inspired by Sketchpad, Ivan Sutherland’s early computer-aided design (CAD) program. Each presents an occasion for thematic exploration and conceptual recombination, revealing particular facets of the medium of computer graphics—its genealogy, techniques, culture, and influence.

Gaboury’s decision to orient the book around discrete yet materially elusive objects isn’t simply a methodological preference; rather, it mimes the monadic theory of objects underpinning the work done in computer graphics at this time, according to which everything in the world is conceived of as an abstract entity available for modeling, simulation, and interaction. The author credits computer graphics for this world-image. He argues that the field’s preoccupation with accurately rendering surfaces, shapes, textures, and lighting entailed a new conception of reality; every object in one’s environment became qualitatively flattened in its construction, dissolving into a set of control points available for manipulation. From this standpoint, objects were not appreciated for their functional or sensuous specificities, but as vehicles for the representation of solutions to problems, neutral surrogates for any object at all.

A modern rendering of the Utah teapot, 2009. Photo: Doug Hatfield. CC BY-SA 3.0

Take the Utah teapot. This object—familiar to designers as a standard reference model in contemporary software suites like Autodesk 3ds Max, LightWave 3D, and AutoCAD, and the focus of an entire chapter in Image Objects—is now an inside joke in the field, its appearance evoking an entire half-century of research, practice, and industry. Based on a simple ceramic Melitta vessel and first modeled by Utah researcher Martin Newell in 1974, it initially served a practical purpose at a time when the department was tired of testing new algorithms on easily generated shapes; researchers wanted new, readily recognizable objects that synthesized then-recent mathematical solutions for parsing complex, irregular curvature, including Bézier curves, Coons patches, NURB surfaces, and b-splines. Newell’s teapot, once digitized and displayed as a wireframe, was instantly popularized through a set of academic papers and, eventually—after a text file containing its patch parameters was shared over the ARPANET network—the precursor to today’s internet.

The ubiquity of the Utah teapot in computer graphics, then as today, testifies to the University of Utah program’s vast web of influence. It also indexes a time in which, via formalized translation methods between tangible and virtual objects, the standards for rendering digital images began to refract back onto the design of the physical commodities now populating our perceptual field. What Gaboury calls the “slipstream aesthetic”—initially attributable to the World War II–era engineering of smooth, aerodynamic bends and arcs found in aircraft, automobiles, and ships—became suffused in all kinds of consumables meant to be gripped by the human hand: The “blob-like appearance of your water bottle or video game controller” are two reflections of it, informed to a great extent by the mediation of graphical techniques. By pushing computing into tighter intimacy with the physical world—away from the execution of procedural calculations in time, toward the assembly of relational objects in space—computer graphics altered the general conceptualization of the modern computer’s functioning, even its purpose. That intimacy always carried within it a profound ambivalence: The simultaneous anticipation and anxiety once evoked by postmodern theory’s simulacrum society and today by VR and the Metaverse are expressions of it. But this new world wrought by computing was implicit in Jurassic Park’s control room: where a language perfected for describing reality was endowed with the capacity to make it.