As I read through David Gelertner’s Mirror Worlds; or the day software puts the universe in a shoe box…how it will happen and what it will mean I cannot help noticing how quickly his exuberant discourse on emerging software designs slides into the rhetoric of the electronic sublime so eloquently described by James Carey and John Quirk over twenty years ago.
I am reminded as well of a recent film.
Wim Wenders’ 1991 film, Until The End Of The World, follows the adventures of Claire Tourneur into the not too distant dystopia future. Although it is set against the backdrop of a possible nuclear disaster, unlike other films which feature a nuclear apocalypse, Wenders eschews the fantastical protocyberpunk reality often visualized in the genre. The world depicted in the film is populated with both familiar remnants of present systems and technologies whose arrival seems imminent. Prominent amongst these are: satellite monitored roads that appear on the car dashboard, miniature videophones that fit into the palm of one’s hand, and cameras which record brain wave patterns the experience of seeing.
All of these technologies facilitate “instantaneous” communication and provide immediate answers to one’s questions irrespective of geographic location. Indeed, the whole process of monitoring and surveillance is a subtext of the film. Yet, the biggest concern of the film’s protagonists is neither the pervasiveness of the web of surveillance for they simultaneously are tracked by it and employ it to track others nor the threat of bodily harm in the wake of a blast, but rather the realization that a nuclear explosion disrupts electrical systems, rendering all the high tech toys they have come to rely on non-functional.
Welcome to mirror worlds.
David Gelertner’s Mirror Worlds is set within a similar future technological possibility minus the threat of a nuclear explosion. Gelertner does not make films. He is an Associate Professor of Computer Science at Yale University whose book cover biography states that his expertise is programming languages and methods, and artificial intelligence. Mirror Worlds explicates current software designs, as well as designs yet to come, which render possible the high tech future captured by Wenders on film.
Mirror Worlds oscillates between detailed descriptions of the conceptual basis common to all state-of-the-art software designs, social and philosophical ruminations on how this will change our lives for the better, and moralistic exhortations to jump on the electronic bandwagon. The appeal of this book, marketed as a science text by Oxford, is that it presents technical information from within a computer science perspective on such topics as software/hardware, artificial intelligence, “tuples,” recursive structures, maps and scripts, modules and procedures, “turingware,” parallel processing etc. etc., in non-technical language. Gelertner asserts that software designers are the civil engineers of our day, “constructing the labyrinths we inhabit and make use of more and more.” (35) As Gelertner never tires of reiterating, as citizens, we should be concerned about these software developments because they concern us. “Worry” is in order, but this is minimal compared to the promised exhilaration upon occupation of this system. (360) While Gelertner’s examination of “what it will mean” is more facile than his analysis of “how it will happen,” Mirror Worlds is a prescient text for those attentive to the social and political power implications potentially embedded in software design.
What is a mirror world? Mirror worlds are “scientific viewing tools” (microscopes, telescopes), focussed on “the human scale world of organizations, institutions, and machines.” (5) According to Gelertner, the images of geographical space that will appear on our future computer screen so wonderfully depicted by Wenders are mirrors of our ‘object world’ that have as their prototypes: peepboxes, zen gardens, and orbs. For Gelertner these diverse phenomena indicate a fundamental and irrepressible trans-cultural human urge to grasp the whole. Computerized renditions of these phenomena, or mirror worlds, are not simulated fantastical realities but a meta-reality, a software model of any location or sublocation whether it be an institution, a city, or a country; Gelertner gleefully imagines the universe made accessible “on screen.” These models will appear because the computer, or rather its software, will be wired up to the everyday world through various “probes or sensors.” Not surprisingly, the number and types of probes needed to sustain such a system is one of the least developed areas in the book, eliding the question of the amount of probing and the nature of the hardware needed to generate the data that he assures the reader will empower us.
As Gelertner envisions it, the software ensembles which coordinate these images on your screen have the potential to show each of us exactly what we want to see. It can sustain a “million different views, a million different focuses on the same city simultaneously.” (6) Each visitor will be able to zoom in, pan, roam through this model as they please, at whatever pace and level of detail is convenient. On departing, we can leave behind a contingent of software alter egos who will keep tabs on whatever interests us. I will return to this idea later. Most important, the software can remember its own history in perfect detail. It has unlimited storage, and a myriad of internal mechanisms for automatically ‘squishing’ (his term not mine) together like data to optimize storage, and it can recall this history when it is asked. All changes in the human world instantaneously appear in the mirror world, which operates in real time. When one turns on the computer and enters into a mirror world, one conceivably enters into a moving, changing image captured in real time through various means of data input.
It is important to understand that for Gelertner, a mirror world is not made possible by computers, which are simply machines that have memory and processing capacities, but through software developments. Gelertner offers some useful insight into the meaning and function of software, explaining that software is not simply a list of instructions one gives to a computer. (9) Software is simultaneously the building material of the information-age, an architectural plan, and an information machine: “hardware guides and transforms forces; electronic machinery does the same with electrical signals; software machines transform information.” (9) The term infomachine is frequently deployed as a euphemism for software. The distinction between software and hardware is one of the most useful and problematic areas in his book.
In a classic Cartesian move, Gelertner sees the computer as body and software as mind: “The program text the words and symbols that the programmer composes, that ‘tell the computer what to do’ is a disembodied information machine. Your computer provides a body,” and “…bodies are a dime a dozen.” (39) A disembodied information machine is a program or plan written down in full of what that machine might do. A running program, which requires a computer to make it accesible to us, is the embodiment of a disembodied machine (software). (39)
To move the analysis to another level, mirror worlds are software replicas of the world, powered by computers, that have the potential to feed not simply streams of data, but oceans of data into an “asynchronically” coordinated software “ensemble”, where this data will be ordered and organized for maximum efficiency, clarity and accessibility. An “asynchronous ensemble” is a group of objects that interact together, but are capable of performing separate functions to maximize efficiency. To use the example Gelertner provides, if you put a hundred toasters together side by side, you have a group of toasters. If you put a hundred monkeys side by side, you have an entity that is more than the sum of its separate parts. (67) Asynchronous ensembles are the “crucial technology of nature and mankind.” (68) Apparently, ensembles are not restricted to animals or humans: “A watch, a steam engine or a factory is built up of many simultaneously active components.” (69) Chemical, biological, and social systems are also described as asynchronous ensembles of one kind or another. In the latest extension of the cybernetic vision of the interchangability of human, animaland machinem, Gelertner imagines that “Software ensembles can be modeled after natural ones…; or they can be models of natural ones…; or they can blend together with natural ensembles.” (68) He describes this blending of our mental capacities with software as “turingware.” The language of technological determinism permeates Gelertner’s text at crucial moments like this: “Ensembles, then, are natural and they are inevitable.” (74)
Ensembles also make speed and uncoupling possible. While contemporary computers work faster and faster, signals travel only so fast down a wire and through a system. Ensembles provide the kind of speed needed for a mirror world to function. (71) “With ensembles, the physical limits of how much information can be carried by a given signal, all but vanish.” (73) These developments supplant the need for bigger and faster computers to manage all this information, because they harness the potential of smaller computers to work on the same problem, reducing processing time through parallel processing. One carries out huge numbers of simple operations on single computers rather than on one megacomputer this disassemblage and reassemblage of instructions and information between computers and softwares is done via “tuples.” While the creation of communication gateways between all software programs has not occurred, we are assured that programmers are working in this direction as you read this paper. In fact, Gelertner’s research group has built a simple ensemble known as “Linda.”
The implications of ensemble building one of the bases of mirror worlds are more profound than the simple amalgamation of all data bases. Mirror worlds are conceived of as prosthetic extensions for seeing, understanding and decision making because they offer us ‘topsight’ a neologism coined by the author. In both software design and human life, Gelertner asserts, the “management of complexity” is the goal. When building programs software designers achieve this by organizing space and time into various sets of maps (modules) and scripts (procedures) according to three fundamental principles: through recursive simplicity, uncoupling and espalier. (54)
An object is recursive when the whole is structurally identical to its parts. Fractals are recursive in nature, and in the case of software, large programs are built out of smaller ones, using the same subdivision of a “landscape” into a “plot.” At the center of each plot is a command post containing an “actor,” or set of procedures, whose job is to respond to information that comes its way for conversion into some other form. “Uncoupling” means pulling apart a complex problem into separate components. This makes using an ensemble of agents a possibility. Subdivisions follow the structure of recursive logic. Finally, data is reintegrated in the form of a ladder that moves from bottom to top, with the raw data being fed into the system at the bottom and a simplified abstraction appearing at the top: this structure recurs in various forms throughout the book. The two predominant forms are as “trellises” and “espaliers.”
While espalier and uncoupling are important concepts, recursivity is key to the entire system. Each larger piece of software (info machine) contains subplots and regions that are other infomachines. It grows and shrinks, it may spiral into a recursive hole and spiral out again. It is a one way mirror in the sense that it can see out, make use of the infochunks and procedures in the surrounding landscape, but the surrounding landscape cannot grab info chunks in each command post. (59) Not only is this a description of the principle upon which software is built, but it is a description of the procedure one uses to search for information and find information in a mirror world. “In this way I can pilot my thought-glider gracefully downwards level-by level towards the nitty gritty, assimilating details gradually, never losing sight of where each detail fits.” (55) Together recursivity, uncoupling and the creation of ensembles allows “organization” or “control” at every level. Recursivity, uncoupling and espalier are the principles by which the author structures his text. Related concepts appear at different points offering more or less detailed explanations of how software works.
Gelertner hints that there are practical applications for mirror worlds: a diminishment of annoyances such as traffic jams; monitoring crime; entering into politics. But why such urgency in the text? He legitimates this massive system with the promise that it will give us topsight, a way to grasp the whole of the lifeworlds, and make it manageable for confused citizens, thereby revitalizing a decaying polis in which we have been reduced to “ants.” Society has become too big and complex to manage or understand, he says. He claims over and over that mirror worlds will help us comprehend the “dangerously complicated and basically indifferent man-made environments that enmesh you, and that control you to the extent that you don’t control them. ” This ant vision…is incompatible with a free society.” (34) He offers this pseudo-option: we could simplify society to make it suit our ant-like vision; or we could “sharpen our vision instead” fight progress with progress with “the wholeness enhancing lenses of our mirror worlds” (34) While we will still be encompassed by the large institutions of modern life, they will also be encompassed by us as long as we have a computer, a modem, a software, and the inclination to use it.
Gelertner naively believes that such a system would give “ordinary citizens” more power vis a vis bureaucratic institutions (and here, of course, he continually uses examples of government rather than corporate control). He is a zealot, making those who are disinterested or refuse equivalent to irresponsible citizens: “you might be such a bored and apathetic goof that you never turn it on, never even peek inside a mirror world.” (34-35) As one of the converted, he overlooks the obvious point that in an analogous example, the new “information highway” being built extends the existing powers of the dominant corporations who are benefiting from the creation of these “services, ” most of which are transactional in nature (money for product), rather than informational. In the latest recommendations from the Canadian Radio Television and Telecommunications Commision’s (CRTC) hearings on convergence among telephone, broadcasting and cable systems, citizens are being asked to foot the bill now for the cable infrastructure which will be built in Canada and later to pay for use of the service. These policy choices are being circumscribed according to the hegemonic interests currently at work: why should it be any different when mirror worlds are implemented? Like many who subscribe to the technological sublime, Gelertner fervently argues that technology itself is an active agent in the service of greater freedoms. More technology is the panacea to rectify social and political problems automatically, ignoring the current context of power or other social and historical information to the contrary.
There is another level at which the implications of future mirror worlds should be addressed. Ironically, the organization of software programs into manageable mirror worlds attempts to shore up the sense of self that is, to give individuals more decisional power at the same time as the self is conceived of as divisible into ever more minute functions. Upon entry into a mirror world, we are offered multiple perspectives, invited to split ourselves to obtain simultaneous points of views: it thus “overcomes” the limits of the current body in which we only ever occupy one place at one time. (5) In the info-landscape of the mirror world we can set up software alter-egos wherever we want. These alter-egos or agents keep tabs on developments in different areas of the “city.” My alter-egos, agents, or virtual “knowbots” will be programmed with my interests, and with speculative powers Gelertner devotes a long section to artificial intelligence and therefore, this “knowbot” will be able to find and filter information custom made for me. In mirror worlds, we simultaneously transcend the world and are immersed in it. It is trans -space and trans-time: Gelertner envisions that you can leave messages for your future self in different locations. As Gelertner says, in this schema, “each software agent is an information machine, using an information machine to build an information filter, to assign each visitor his/her own separate information machine as a guide.” (74) This is a conception of the split subject as an information based self that is more fractured than ever imagined. Not only are mirror worlds recursive, uncoupled and trellised in nature, but the conception of the self we are offered is a recursive self right down to the level of our DNA, which after all, is conceived of as a piece of coded information, phylo and onto genetically passed on.
Gelertner’s assertion that “the social implications of these software gizmos make them far too important to be left in the hands of the computer sciencarchy,” (5) is correct. However, Gelertner does not adequately explore the social implications of this software; he reduces it to a battle between technocratic realism and a nostalgic neo-romantic longing for the past, in his last chapter.
In the classic panoptic system described by Michel Foucault in Surveiller Et Punir, a single person could control the entire system by placing him or herself at the center. The outside was always illuminated, hence visible and subject to control despite the imbalance of numbers. In the post-panoptic system, the power relations are much more destabilized and dynamic, but no less effective. We are told, “when you enter a mirror world you enter with a cameraman by your side.” (196) At the same time, one realizes that the system is structured so that everyone else has a camera. Under the three cardinal principles of recursive simplicity, uncoupling and trellis building, Mirror Worlds captures the architecture of visibility and social control in a post-panoptic culture. Some of its features are as follows:
The system is radically decentralized, there is no central guard who monitors the prisoners on the outside, indeed, there are no walls, but only a network, or asynchronous ensembles. Yet, this decentralization only further serves to render the operation of power more efficiently and intensively.
We simultaneously occupy multiple positions within this system, rupturing most clear dialectical distinctions between oppressed and oppressor; we set up agents or spies, enter with cameramen by our sides, at the same time as we are obviously a part of any scenario being depicted.
It is not static, and doesn’t operate through accumulation, but through continual re-investment and circulation “nothing collects in stagnant pools…Everything flows.” (137)
At the same time that there is great flexibility and multiple positioning, its structure is corporatist; there is a hierarchy of information and beings, and a logic of fragmentation (uncoupling), and a system of values.
It preserves the Cartesian dualism of mind over matter, and its related biases, particularly as they relate to the gender split.
It is trans-space and trans-time: You can talk with anyone anywhere, ‘leave messages for your future self,’ and have instant access to any past information stored in the system.
It is attached to all forms of bio-organic and non-bio-organic being, and in fact, breaks down the distinction between software and human beings in a continuation of the original cybernetic dream of identifying control mechanisms in animals, humans, or machines.
It translates these monitored movements into productive data that can be used to generate more knowledge about both the subjects actively plugged into the system and those who are not. Passive and active surveying is integral to the system.
While not solely determined by one variable (the economic), it nevertheless collapses the political into the economic registers. Our identity as citizens is replaced by the notion of a client.
It operates under the rhetorical guise of democracy, greater choice, and freedom, as it aims for efficiency and the management of complexity.
Finally, at the same time as the system works through absolute imminence, and aims for total inclusion, it promises transcendence, and believes that this is possible through more data collection. It is tautological in this respect, but never asks why this mirror world should be purified of the social, political and personal malaises of the non-mirror world.
After carefully considering all of this, I ask: Do I really want the universe to be put in a shoebox? Is ‘topsight’ really that critical, or is this just a more sophisticated utilitarian toy for making assessments divorced from a discussion of ethical ends, a toy whose final equation is not more topsight for ants, but greater and more diffuse social control? Even if more information services are the result, who will have access to what information in any future mirror worlds? Will the software be readily available to all? Is this technolgical realism, or the latest incarnation of what C. Wright Mills termed “crack pot realism?”
In Wender’s film, Claire is brought to a place where a scientist is experimenting with the recording of dreams, captured and replayed on miniature monitors. The dreams on screen are beautiful. You can see yourself as you once were, and as you imagine yourself to be. But there is one drawback they are addictive. Once you begin to watch you lose the desire to interact with others, you lose the desire to eat, you cannot tear yourself away and you neglect your corporeal body. Claire becomes addicted to her own dreams. When her batteries are taken away, and it appears she will die from image-withdrawal, she is cured by being reintroduced to story telling and print; it breaks the solipsistic circuit that has entrapped her. She gets a job on a space station monitoring corporations and governments for oceanic pollution crimes.
I don’t know whether mirror worlds will take us to “the end of the world.” I don’t know if I agree with Wender’s analysis of the addictive nature of images, or his simple text-cure. I do know that his film helped me to visualize David Gelertner’s text-based diagram of the mirror worlds he imagines will be contained in “pint-sized” computers. The book, Mirror Worlds, doesn’t adequately explore what the implementation of such a system will mean, although it provides invaluable information on “how” it will be put together. For future travelers being cajoled into buying a ticket on the bus moving along the information highway, MIRROR WORLDS is essential reading.