The Architecture of Information: Open Source Software and Tactical Poststructuralist Anarchism
Open Source Software refers to a software development model in which the source code is open for modification and redistribution, unlike proprietary software such as Microsoft Windows, which denies access to the source. The OSS model, in particular the Linux operating system, has garnered much attention from disciplines as diverse as computer science, sociology, economics, law, and political science; however, cultural theory and media studies, especially theories influenced by poststructuralist thought, have yet to address the social impact of Open Source and its potential as a political philosophy in the network society. This paper examines the convergence of poststructuralist anarchism (using works by Michel Foucault, Gilles Deleuze, Lebbeus Woods, and Hakim Bey) and Open Source Software via a discursive analysis of Eric Raymond’s Open Source manifesto and ethnographic survey, “The Cathedral and the Bazaar.” —mt
The traces of power in the network society are equally located in the architecture of bricks and mortar and the architecture of information, the discursive practices that constitute the coding of network topologies. This paper examines the discourse of computer programming through Eric Raymond’s ethnographical account of the Open Source software developmental model. Raymond’s Open Source software manifesto, “The Cathedral and the Bazaar” (hereafter, “CatB”), the infamous text that inspired Netscape to release the source code for its Web browser in 1998 in an attempt to compete with Microsoft’s Internet Explorer, differentiates Open Source software development, which is figured as nonlinear and self-organizing, from Closed Source, which is represented as hierarchical and authoritarian. The Open Source model has been characterized by some as representing a liberatory politics for the information age. Open Source, as represented in the work of Raymond, is a tactical political philosophy whose central architectural metaphors—the bazaar, and its supplementary term, the cathedral—share theoretical homologies with anarchistic poststructuralist statements by Michel Foucault, Gilles Deleuze, Lebbeus Woods, and Hakim Bey. The intent of this essay is to examine points of generative convergence between the history and philosophy of software development and certain proponents of poststructuralist thought. In doing so, I wish to initiate a cultural analysis of the historically situated discourse that shapes software development, the texts and practices that inform the coding of the architecture of information. Raymond’s bazaar shares theoretical homologies with Deleuze’s modulations, Woods’s freespaces and heterarchic societies, and Bey’s Temporary Autonomous Zone, but it is in the figure of the programmer especially that the intersecting discourses of electronic and concrete spaces converge and the inescapably cultural base of technology emerges.1 The bazaar programmer loses the distinction of “developer” and becomes a “co-developer,” dissolving the categorical distinction between those who code and those for whom there is a Code.
The Discourse of Software Engineering
Critiques of the information age and the role of cyberspace in the operation of the surveillance society often ignore the fact that “electronic space is embedded in the larger dynamics organizing society” (Sassen 1). Cyberspace is not a separate space, a virtual room of its own, but an embodied architecture of networks and actors situated in disparate geographical, sociological, psychological, and economic spaces. The very history of the Internet and its cyberspatial enclaves is immersed in ideologically antagonistic contexts: the Internet was ostensibly developed by the U.S. Department of Defense but nurtured by the libertarian leanings of early hacker culture. In its originary moment, then, the Internet was emblematic of both control and freedom, the apotheosis of the surveillance society and the dream of anarchistic autonomy.2 In Saskia Sassen’s re-theorization of electronic space, the increasing commercialization of public networks and centralization of power in private networks have created “cyber-segmentations” that are contrary to the Internet’s supposed dream of decentralization and openness (2). The encroachment of multinational postindustrial capitalism, the context in which cyberspace is “embedded,” produces online inequalities centralized in the offline spaces of power (5).
Sassen believes that “a focus on place and infrastructure in the new global information economy creates a conceptual and practical opening for questions about the embeddedness of electronic space” (5–6). Taking a cue from Sassen, the current essay focuses on network infrastructure through the metaphors of architecture; the architecture of places and information; urban design and systems design; and the trace of spatialized power that informs them. Cyberspace is constructed by operating systems and software applications. People often overlook the social conditions of software creation, just as they often ignore the cultural conditions of cyberspace, preferring instead to see cyberspace as the digital perfection of our messy analog desires. The “architects of information” engineer software,3 and just as discourse implicates these architects of information in the latent discursive practices of their particular time and place (in this case, much of the ethos of programmer culture emerges from the West Coast of the United States), the software they create registers in the web of normative practices that constitute discourse. A focus on “place and infrastructure,” as suggested by Sassen, must include a focus on the infrastructure of the code, the architecture of information, which contributes as much to the centralization of power in places as the commercialization of access to information, the emergence of infomediaries, and the prevalence of private corporate networks, Sassen’s primary forms of “cyber-segmentation” (9).
If You Build It, They Will Come: Architecture and Ideology
Architect Lebbeus Woods, whose “anarchitecture” attempts to create “freespaces” through indeterminate structures, identifies the relationship between architecture and ideology: “While architects speak of designing space that satisfies human needs, it is actually human needs that are being shaped to satisfy designed space and the abstract systems of thought and organization on which design is based” (279). Mark Poster makes a similar claim for information architects when he discusses the impact of the Superpanopticon: “Surveillance by means of digitally encoded information constitutes new subjects by the language employed in databases” (Mode 94). Architecture, whether it refers to buildings or databases, constructs subject positions through the spatialization of power. Much has been written about the architecture of bricks and mortar, but very little, if anything, has been written about the architecture of information, the discourse of computer programming; and this is so even as the two architectures become inseparable.4 This essay engages a variety of discourses on the sometimes liberatory architectural models of what Todd May calls “tactical poststructuralist anarchism.” In particular, the model of the software engineer as a subject working within and creating autonomous spaces, a tactical anarchism for the information architect represented by CatB, offers an instance of resistance that defies the monolithic rendering of the surveillance society described by Poster and others, an almost cyborgian moment in which information architects “seize the tools that write them” for the purpose of a tactical, temporary “uprising,” to use Hakim Bey’s preferred term.
A Rebel Without A Cause
“Linux is subversive.” One can only wonder if CatB would have had as much success without its provocative first three words. Raymond, described as “the armchair philosopher of the open source world” and “a hard-core libertarian” (Wayner 106–07), originally wrote CatB to demarcate the factions that existed within the Free Software movement and to establish his camp—the more market-friendly, “open source” camp—as the desirable, though not singular, option. For Raymond, Open Source software development offers significant technical benefits over Closed Source, and this improved product development cycle should be the central motivation behind adopting such practices. However, he does not advocate the prohibition of proprietary software. CatB, which was originally presented to the Linux Kongress on 21 May 1997, convinced Netscape to release its Navigator source code on 22 January 1998, an event that brought much public attention to Raymond’s essay and much private-sector approval to Linux, the enormously successful operating system on which CatB is based. Just how subversive Linux is—politically, philosophically, economically, sociologically—remains unclear, but the architectural metaphors on which Raymond’s essay is based may be considered an exemplary expression of techno-cultural anarchistic politics. The advance of Open Source software development has been described as a modern-day Lutheranism, and Raymond’s Open Source manifesto has been described by FEED magazine’s Steven Johnson as “the Port Huron Statement of the movement.” Raymond told Johnson in an email that what Open Source subverts specifically are “classical software-engineering assumptions about how to do high-quality software. More generally, [Open Source subverts] assumptions about the organization of cooperative work—that it has to be organized on either command-and-control or scarcity-economics principles” (“Whole Web”).5 To which Johnson wondered, “Is Open Source the ultimate expression of the techno-libertarians, or a startling revival of the progressive, anti-corporate sentiments that last flourished with the SDS?” Raymond’s brand of Open Source is certainly not anti-corporate, but it does possess a subversive political philosophy, a philosophy I shall categorize here as “tactical poststructuralist anarchism,” embodied by the architectural metaphors of the cathedral and the bazaar. The architectural metaphor foregrounds not only software engineering as a “situated knowledge,” but also the transformation in software engineering enabled by the emergence of the Internet; that is, the infrastructure of the Internet that evolved in the “real” world enabled a reformulation of the political terms in which the very same technological infrastructure could evolve (primarily, as Raymond argues, in the way the Internet enables “parallel debugging”); the architecture of networked computing created a new space for information architecture (the coding of the code).
Use The Source, Luke
In CatB, Raymond employs his titular metaphor to differentiate his brand of Free Software (Open Source) from other forms of Free Software, such as Richard Stallman’s Free Software Foundation (FSF). Raymond believed that the FSF operated too much like cathedral-style development, with a hierarchical structure of authority in which one central project leader (in this case, Stallman) determined the development path for the source. When Raymond became aware of Linux, a Unix-based operating system created by 21-year-old computer science student Linus Torvalds at the University of Helsinki in 1991, he was immensely impressed by the way it had evolved through an open distribution of the source code over the Internet. This decentralized model of software development took advantage of the collective debugging power of thousands of computer programmers and essentially enlisted “users” as “co-developers.” The first official release of Linux came in 1994, with changes being made and new versions being posted daily. Even though Linux was developed five years after Microsoft began developing Windows NT, Linux has become a competitive alternative to NT with a worldwide user base around twelve million (Moore), a fact that many suggest proves the efficacy of its open development model. A recent Business Week cover story on Linux notes the operating system’s accelerated penetration into the server market, from almost no percentage of the market three years ago to 13.7% of the $50.9 billion market in 2002, a figure expected to reach 25.2% by 2006. For this reason, business experts are calling Linux “the biggest threat to Microsoft’s hegemony since the Netscape browser in 1995” (Kerstetter et. al. 78).6 In the short-term this may, however, be an overstatement. Microsoft revealed for the first time in November 2002 that four of its seven business units are losing money, while its Windows division generates an 85% profit margin. In other words, the Windows OS is still where Microsoft makes most of its money, and it is on the desktop where Open Source projects like Linux have made the least impact. That said, Linux and other Open Source software developmental models have proved to be an effective and widely accepted alternative to various proprietary software.7
The Mythical Man-Month
CatB borrows much of its structure and codification of the principles of software engineering from a classic text in the field, The Mythical Man-Month by Frederick P. Brooks. A twentieth-anniversary edition of Brooks’s text was issued just prior to the composition of CatB, and Raymond makes specific mention of the text in his notes. The central thesis of Brooks’s classic text has become known as “Brooks’s Law”: “Adding manpower to a late software project makes it later” (25). Brooks formulated this law before the age of the Internet, before the advent of a truly scalable form of “parallel debugging.” Linus Torvalds was the first person to refute Brooks’ Law in a demonstrably successful application of what Raymond in CatB famously dubs “Linus’ Law”: “Given enough eyeballs, all bugs are shallow.” Adding manpower by distributing the source code freely over the Internet, as Torvalds did in 1994 just as the Internet was emerging as a cultural phenomenon, can actually improve the quality of a piece of software while maintaining its architectural integrity, and it can do so on time. The focus of Raymond’s manifesto, in an intertextual dialogue with Brooks’s software engineering classic, is not the quality of Torvalds’s code in the Linux kernel, “but rather [Linus’s] invention of the Linux development model” (Raymond).
Raymond also derives the cathedral metaphor from Brooks. In The Mythical Man-Month, the cathedral-builder represents the ideal form of software engineering, an architectural unity of design across generations of builders. However, most cathedrals are problematic, says Brooks, because they bear witness to different architects and different styles that have been applied, a visible legacy of heterogeneity.
It is better to have a system omit certain anomalous features and improvements, but to reflect one set of design ideas, than to have one that contains many good but independent and uncoordinated ideas.(42)
It is important to note that when Brooks refers to the architecture of a system, he means “the complete and detailed specification of the user interface,” something not to be confused with the implementation of these design specifications (45). The system architects, then, are something of an aristocracy in his eyes, granted the noble task of envisioning the system in its entirety, authors of the metanarrative, or autocrats of information. Brooks points to the triumph of the WIMP (Windows, Icons, Menus, Pointing) interface as an example of perfect architectural integrity in software engineering, an exemplar of his cathedral-builder.
The WIMP is a superb example of a user interface that has conceptual integrity, achieved by the adoption of a familiar mental model, the desktop metaphor, and its careful consistent extension to exploit a computer graphics implementation.(260)
Of course, the WIMP interface, which found its way from Xerox to Apple to Microsoft, is now the controlling metaphor of the most popular operating system in the world, and perhaps the emblem of proprietary software and the monopolistic practices of the leading cathedral-builder, Microsoft. Windows may be “a superb example of a user interface that has conceptual integrity,” but it is also a grisly parody of the way in which the computer is always mediated by the space of work (in this case, the central metaphors Windows uses to organize the computer’s content and functionality are metaphors of the workplace: trash cans, files, desktops, and so forth). Brooks’s sense of conceptual integrity becomes a sort of normative principle in computer-mediated communication: the desktop metaphor works because it is “a familiar mental model,” but it is a familiar mental model because it reifies the spatial order of hegemonic discourse. If the system’s architecture is more important than incremental but heterogeneous improvements on that design, then where is the space for adaptation? Or, if one conceives of “the code” as a discourse of subjugation, where is the space for counter-hegemonic discourses, what Steven Johnson aptly calls “user-hostile design” (Interface Culture 227)?
J. Mitchell Morse describes cathedral building (the brick-and-mortar kind) as an aspect of “the grandiosity-glorification syndrome, an aspect of the fully developed leisure-class authoritarian syndrome,” and, pointing to texts as diverse as Vico’s New Science and Kafka’s “The Great Wall of China,” he characterizes such stories as tales of “secret government” of “an ancient authoritarian habit that afflicts every nation” (104). He also recalls, citing Vico, that many of these ancient societies built their “secret governments” on the foundations of languages, languages—from hieroglyphs to Latin Bibles—indecipherable to the slavish masses who constructed their pyramids, great walls, and cathedrals. The process of governing was “secret” in the sense that these secret languages reinforced the status of the ruling class without the subjugated entirely understanding the terms of their subjugation; in the cathedral, then, resides a form of “false consciousness.” The cathedral-builders of postmodernity have again harnessed a language at once complex and capable of being “hidden” from the masses: the compiled and proprietary source code. The Windows operating system—whose development is driven by market forces and the corporate interests of Microsoft, perhaps the most powerful cathedral-builder in history because of its global imperative—is the monument of Arthur Kroker’s “virtual class,” the “authoritarian habit” of the ruling class of the information age.
Raymond contrasts the cathedral, an authoritarian and linear model of software development associated by implication with the monopolistic behemoth Microsoft, with the bazaar, an open and nonlinear model of software development whose champion is Linus Torvalds and the Linux operating system. Raymond describes the Linux community as “a great babbling bazaar of differing agendas and approaches . . . out of which a coherent and stable system could seemingly emerge only by a succession of miracles.” As Ko Kuwabara depicts the Open Source community, based largely on Raymond’s ethnographic essays, it behaves much like an emergent system as described by complexity theory. Both Kuwabara and Raymond suggest Linux’s development model succeeded because it obeys certain natural laws governing the evolutionary development of complex systems: “The case of Linux suggests . . . that aggregate behaviors are better understood as emergent properties—properties that emerge out of micro-level interactions among constituents—than as qualities intrinsic to the system and independent of its constituent parts” (Kuwabara). Raymond further compares the Linux community to “a free market or an ecology,” an echo of his technolibertarian ideals; but this representation of “central planning” as the artificial and oppressive antithesis of the more organic and natural evolution of complex systems is too reductive and ignores both certain benefits of central planning and the fact that even Microsoft utilizes Open Source development models some of the time.8 Reductive rhetoric about the natural evolution of “free markets” is especially problematic for those who profit from the Internet because it was originally established and developed by the government. There is a naturalistic fallacy at work here, a troublesome tendency to say the Open Source model is more “natural”—and therefore better—than closed source.9 The bazaar might be better defined by the “tactical poststructuralist anarchism” I discuss below.
First, it should be noted that the primary distinction Raymond makes between the cathedral-builder and the bazaar is the use of “parallel editing.” Raymond writes:
Here, I think, is the core difference underlying the cathedral-building and bazaar styles. In the cathedral-builder view of programming, bugs and development problems are tricky, insidious, deep phenomena. It takes months of scrutiny by a dedicated few to develop confidence that you’ve winkled them all out.
In the bazaar view, on the other hand, you assume that bugs are generally shallow phenomena—or, at least, that they turn shallow pretty quick when exposed to a thousand eager co-developers pounding on every single new release.
This is the essence of Linus’ Law (“given enough eyeballs, all bugs are shallow”), and it exemplifies the means by which Open Source deconstructs the otherwise stable and antithetical subject positions of “developer” and “user.” According to Linus’ Law, a “user,” formerly the colonized subject of imperialist cathedral-builders, now participates in the development of the code and is an equal “owner” of the code. Instead of “users” and “developers,” everyone is a potential “co-developer.” The bazaar becomes a liminal zone of subject-creation, an interstitial space for the negotiation of identity.
Homi Bhabba, commenting on Clifford Geertz’s use of the bazaar and the English gentlemen’s club as social spaces of diversity and homogeneity respectively, imagines a more “agonistic and ambivalent” relationship between the two, suggesting that “anxious passage” between the two merges “the encounter between the ontological cultural impulse and the memory of the displacements that make national cultures possible” (36). The bazaar has traditionally been identified with oriental exoticism; however, recent studies refute this colonialist historicism and reveal that bazaar markets “can equally be found in developed industrialised economies” (Fansellow 262). The use of the “bazaar” as a metaphor for Open Source immediately performs a sense of otherness and dislocation, but it does not establish Open Source as the binary opposite of cathedral building; while the two might be construed as equivalent terms of comparison—one evoking Christianity, the other Islam—Raymond’s stated intention when employing the “bazaar” metaphor was to invoke the marketplace, not religion. He was originally going to call the essay “The Cathedral and the Agora,” to suggest an open space of public exchange. The convergence of the metaphors occurs on the level of “enclosed” and “open” spaces, the cathedral the epitome of an enclosed space, and the bazaar an open space that for some economists is the “model of the perfectly competitive market” with “the apparent absence of monopolistic competition” (Fansellow 262).
The second central feature of the bazaar model is its adaptive teleology. While Raymond requires some sort of goal for the development community (“When you start community-building, what you need to be able to present is a plausible premise”), it is not a fixed goal. Rather, what makes the Linux model advantageous is its ability to adapt to shifting contexts, a tactical philosophy that contrasts with the strategic philosophy of cathedral-building. This tactical approach “produces a self-correcting spontaneous order more elaborate and efficient than any amount of central planning could have achieved” (Raymond). The adaptive order of the bazaar, with its tactical constitution of “the code,” resembles the poststructuralist construction of the subject through cultural codes and a multiplicity of possible subject positions. This analogy is developed in greater detail below.
Tactical Poststructuralist Anarchism
To this point, I have discussed the representation of Open Source software development in Eric Raymond’s “The Cathedral and the Bazaar,” a seminal work in the public dissemination of hacker culture. Raymond’s essay is an inescapably ideological work, from the provocative opening sentence to the engaging architectural metaphors. The attributes of the bazaar software development model share theoretical homologies with what Todd May calls “tactical poststructuralist anarchism,” and an examination of some prominent forms of poststructuralist anarchism will provide the necessary bridge in this analysis from the discourse of software engineering to the political philosophy of poststructuralist anarchism.
In the introduction to The Political Philosophy of Poststructuralist Anarchism, May attempts to map the migratory space that political philosophy inhabits, and he suggests the tension between metaphysics and ethics (essentially between what is and what ought to be) determines the mutable boundaries of this discursive site. He divides the spectrum of political philosophies into the categories of formal, strategic, and tactical. Formal political philosophy cleaves “either to the pole of what ought to be or to the pole of what is at the expense of the tension between the two” (4). Strategic political philosophy recognizes the contingencies of history and social conditions, often addressing the concrete historical conditions under which the political “strategies” are composed, but never allowing ethical goals to be subsumed by contextualizations; thus the aim is to promote the “assumed justice,” the ethical goals, given the historical context (8). Strategic philosophies graph power onto the world in the form of “concentric circles” which emanate from a central “core problem” (10). This concept of power distinguishes strategic political philosophies from tactical, since tactical philosophies, such as anarchism and elements of French poststructuralism, May argues, remain “in the tension,” where there is no center of power, only intersecting practices of power (11). By remaining in the tension between what ought to be and what is, unlike formal and strategic philosophies which tend to gravitate to one pole or the other, and by conceptualizing power spatially as a network of interconnecting matrices, tactical philosophies reject representational political intervention, such as that of a vanguard, and embrace the decentralization or removal of hierarchical power structures. May believes the move from macropolitics to micropolitics in French poststructuralism, in particular the work of Foucault, Deleuze, and early Lyotard, is analogous in many ways to the tactical philosophy of many early and modern anarchists.
The bazaar model of software development represents a tactical political philosophy through its emphasis on decentralized authority, its spatialization of power dispersed in the networks of the Internet, and its rejection of political intervention through an overtly libertarian endorsement of a free-market philosophy.10 Perhaps most important, the bazaar model emphasizes its adaptability to transient conditions through extensive parallel debugging; so, unlike formal or strategic political philosophies, the bazaar model encourages tactical interventions over strategic, teleological planning. In the following sections, I examine the theoretical homologies the bazaar model shares with other poststructuralist tactical interventions: Gilles Deleuze’s critique of Foucault’s Panopticism, Lebbeus Woods’s “anarchitecture,” and Hakim Bey’s Temporary Autonomous Zone. In each case, the tactical anarchism depends on architecture to resolve the contingent boundaries of electronic space as electronic space is embedded in techno-cultural narratives, and this translation of anarchistic poststructuralist thinking via architecture suggests an emerging discourse for the subject positions of the architects of information in postmodernity.
The Mode of Information
The new linguistic experiences of the late twentieth century have been termed “the mode of information” by Mark Poster. Poster focuses on Foucault’s discussion of the Panopticon, an architectural innovation of the nineteenth century representative of the transformation from sovereign to disciplinary power in Western societies. The Panopticon, Foucault believed, epitomized a technology of power, a general model for power, “a way of defining power relations in terms of the everyday life of men” (Discipline 205). The efficiency of panoptic power derives from its ability to automize and disindividualize its power (202). Foucault writes:
The efficiency of power, its constraining force have, in a sense, passed over to the other side—to the side of its surface of application. He who is subjected to a field of visibility, and who knows it, assumes responsibility for the constraints of power; he makes them play spontaneously upon himself; he inscribes in himself the power relation in which he simultaneously plays both roles; he becomes the principle of his own subjection.(202–03)
Poster demonstrates how panopticism as a technology of power operates within the twentieth century in the form of state bureaucracy and computer technology. Both the Panopticon as an architectural innovation, and bureaucracy and computer technology, foster the principles of disciplinary control. “Indeed,” writes Poster in Foucault, Marxism & History,
they expand its scope to a new level. With the mechanisms of information processing (the bureaucracy using people; the computer using machines), the ability to monitor behavior is extended considerably. The techniques of discipline no longer need rely on methods of regulating bodies in space as Foucault thinks. In the electronic age, spatial limitations are bypassed as restraints on the controlling hierarchies. All that is needed are traces of behavior; credit card activity, traffic tickets, telephone bills, loan applications, welfare files, fingerprints, income transactions, library records, and so forth.(103)
Disciplinary control in the information age surpasses spatial and temporal limitations through the advances in technology: “Today’s ‘circuits of communication’ and the databases they generate constitute a Superpanopticon, a system of surveillance without walls, windows, towers or guards. The quantitative advances in the technologies of surveillance result in a qualitative change in the microphysics of power” (Poster, Mode 93). Not only does the Superpanopticon erase spatial and temporal restrictions through the immediacy of information exchange, it also challenges the social hierarchy by circumventing the existing methods of domination: “In the mode of information the subject is no longer located in a point in absolute time/space, enjoying a physical, fixed vantage point from which rationally to calculate its options” (15).
Open Source may represent a space of resistance for the surveillance society, and the metaphor of the bazaar suggests the mode of resistance. In his critique of Foucault’s panoptic technology of power, Gilles Deleuze contrasts the “spaces of enclosure” in disciplinary societies—Foucault’s hospitals, prisons, and factories—with the “modulations” of societies of control, the ephemeral corporations and databanks that organize contemporary life. Deleuze describes the continuous becoming of the subject in the society of control:
Enclosures are molds, distinct castings, but controls are a modulation, like a self-deforming cast that will continuously change from one moment to the other, or like a sieve whose mesh will transmute from point to point.(4)
In the disciplinary society, resistance is localized, the source of power in the maintenance of molds, fixed spaces. In the control society, power is distributed and adaptive, morphing ceaselessly to the contours of resistance. The cathedral-builder prizes conceptual integrity across generations; the bazaar model adapts to contexts without being led by market forces or corporate visions. The bazaar model is a particularly compelling political philosophy because it translates the architecture of enclosures to the architecture of information, and information, the code, mediates power in the control society.
In the societies of control . . . what is important is no longer either a signature or a number, but a code: the code is a password, while on the other hand the disciplinary societies are regulated by watchwords (as much from the point of view of integration as from that of resistance). The numerical language of control is made of codes that mark access to information, or reject it.(5)
The “numerical language of control,” the parlance of information architects, reaches its apotheosis in the biotech century with the coding of the Human Genome Project, “the code of codes” (Haraway 245). The human genome could represent the metapassword, access to unparalleled control or anarchic freedom. Much of the constitution and subjugation of the subject under this regime is a cultural product of database architecture:
Computerized database design is at the leading edge of genomics research. Design decisions about these huge databases shape what can be easily compared to what else, and so determine the kinds of users that can be made of the original data. Such decisions structure the kinds of ideas of the species that can be sustained.(Haraway 247)
Open Source software development, the free distribution of source code for unlimited revisions, carries serious ontological consequences under this regime, consequences that go beyond the scope of this article. The bazaar becomes, quite literally, a site for the construction and reconstruction of the subject, a potential space for autonomous agents to recontextualize the configurations of power in the networks of the control and surveillance society. The architecture of enclosure gives way to the architecture of universal encoding.
The burgeoning availability and application of Open Source software and its development model, at a time when the “mode of information” and its attendant means of surveillance and control are the dominant discourse, resemble the cultural conditions of what has been dubbed the “Foucault paradox,” a situation in which the emergence of the hegemonic discourse of power being described by Foucault emerges concomitantly with its counter-hegemonic discourse:
In other words, the burgeoning panopticism of nineteenth century institutions emerged hand-in-hand with collective discourses of social rights. They bore a reciprocal relation to one another.(Lyon 672)
It is not surprising that the conditions in which the bazaar model emerges are the conditions in which the cathedral model is dominant. Arguably, Microsoft created the personal computing industry through the widespread adoption of an easy-to-use operating system for non-expert users. Microsoft, through its development of the personal computing market, created subsets of software industries that would not exist without the widespread adoption of its operating system. Without this enormous base of users, there would be no need for the employment of programmers in great numbers, and the code produced would not have the same influence over a large number of users. The number of willing “co-developers” available for Open Source projects would also have been greatly reduced without the home computing revolution created by Microsoft. The “Foucault paradox” informs the twentieth century as it did the nineteenth, and increased attention to these co-emergent phenomena would produce more balanced evaluations of the surveillance society and the resistant spaces within.
Lebbeus Woods’s Radical Reconstruction
The play of autonomous spaces in the architecture of information through the implementation of the bazaar model finds its analogous development in poststructuralist theorizing of space with architect Lebbeus Woods’s notion of “freespace.” Woods proposes that normal architecture prescribes intersubjectivity: “Design is a means of controlling human behavior, and of maintaining this control into the future” (279). Like the cathedral, normative architecture constructs object relations in space that precede and prescribe the constituent agents in the network of social relations: “People come and go, ways of living change, but architecture endures, an idealization of living” (281). Brooks praises this form of architecture in The Mythical Man-Month, but information technology has altered the ideological role of space: the enclosures of discipline no longer suffice. Woods proposes freespaces as spaces with potential for functionality but no prescriptive function; as a consequence, freespaces do not impose behavior. Woods introduced the concept of freespace with the Berlin Free-Zone project, what he calls “a hidden city within the one now being shaped” (286). The Berlin Free-Zone embodies the networked subjectivity of Open Source, the free play of signifiers in the space of computer-mediated communication:
This hidden city is called a free-zone, because it provides unlimited free access to communications and to other, more esoteric, networks at present reserved for the major institutions of government and commerce—but also because interaction and dialogue are unrestricted conventions of behavior enforced by these institutions.(Woods 286)
Woods wishes to alter the role of architecture from a discipline that reifies the social hierarchy to a discipline that enables “heterarchy,” the rule of many. “Heterarchical urban forms” such as the Berlin Free-Zone
are invented in response to increasing emphasis in the present culture on the idea of the “the individual,” coupled with recent technological developments, such as the personal computer and systems of communication that simultaneously weaken the established hierarchies by accessing the information formerly controlled by them, and strengthen the autonomy of individuals.(288)
The heterarchical form of Open Source’s information architecture certainly weakens “the established hierarchies,” namely the cathedral-builders, and returns an increased degree of information access to individuals, the legion of co-developers. Freespace is not new, but its public implementation is, says Woods. Similarly, free source code is not new; the information superhighway was paved with the asphalt of free code. But with Microsoft’s monopolistic hold on the productivity software industry, and with a very public battle with the U.S. Department of Justice revealing Microsoft’s anti-competitive practices to the public, the Open Source movement has accrued momentum and become a public emblem of “heterarchic” software development, software without ends.
The Temporary Autonomous Zone
The “mode of information” measures the linguistic experiences of the surveillance society, and the “anarchitecture” of Woods constructs freespaces for the autonomy of the individual in ideologically suffused spaces. Open Source creates an anarchic space for tactical intervention in the surveillance and control society by making the principal means of control, the code, “visible” to the greatest number of subjects.11 The subject, though participant in its own self-subjugation, is also a participant in its own emancipation. The machinations of surveillance, the operating systems and applications of subject construction, are potentially exposed and reconstructed. Always, whether the Superpanopticon’s architecture of information or Woods’s informed architect, the slippage between concrete and virtual architecture, between buildings and databases, leaves traces of the embeddedness of architectural discourse in the web of technocultural discourses. Perhaps the consummate expression of tactical poststructuralist anarchism and its polyglot of technological, scientific, architectural, economic, and political discourses is Hakim Bey’s Temporary Autonomous Zone (TAZ).
The TAZ is a conceptual configuration of the bazaar as political philosophy. Like the bazaar, the TAZ values transient uprisings over sustained development. Like the bazaar, the TAZ is a freespace for individual autonomy, a state prior to mediation. And like the bazaar, the TAZ is a metaphor for the spatialization of liberation, “a microcosm of that ‘anarchist dream’ of a free culture” (Bey). Unlike the cathedral-builders, who must maintain conceptual integrity across generations, to serve corporate interests, even when it no longer proves adaptive to the environment, the TAZ (and now by extension, the bazaar) forms and reforms without teleological motivations:
The TAZ is like an uprising that does not engage directly with the State, a guerilla operation which liberates an area (of land, of time, of imagination) and then dissolves itself to re-form elsewhere/elsewhen, before the State can crush it.
This is not a revolutionary paradigm with a vanguard and a utopian dream of a classless society. The TAZ is a revolution that fails, but only because success would be the ultimate failure, the denial of future TAZs. In the society of control based on surveillance and simulacrum, where the State is—in the true sense of the Superpanopticon—all-seeing, the TAZ is an enclave of the real that cannot be mapped, simulated, made spectacle. “As soon as the TAZ is named (represented, mediated), it must vanish, it will vanish, leaving behind it an empty husk, only to spring up again somewhere else, once again invisible because undefinable in terms of the Spectacle” (Bey). While the Superpanopticon makes space visible for surveillance, mapping, and control, the TAZ will only exist in open spaces like Woods’s freespaces, spaces without ideological impressions:
We are looking for “spaces” (geographic, social, cultural, imaginal) with potential to flower as autonomous zones—and we are looking for times in which these spaces are relatively open, either through neglect on the part of the State or because they have somehow escaped notice by the mapmakers, or for whatever reason.
Bey believes there are times and places in history when TAZs existed, and still times and places where they may exist. Bey uses pirates as historical examples of an enclave of freespace that for a time evaded the “map” of authority, temporary instantiations of freedom. The nomadic figure of the pirate is one in a series of nomads, whose “psychic nomadism” Bey refers to as “a tactic”:
These nomads practice the razzia, they are corsairs, they are viruses; they have both need and desire for TAZs, camps of black tents under the desert stars, interzones, hidden fortified oases along secret caravan routes, “liberated” bits of jungle and bad-land, no-go areas, black markets, and underground bazaars.
Again, the image of the bazaar surfaces, except this time it represents a congerie of spatial convergences in Bey’s “psychotopology of everyday life,” the rootless desires and tactics of the cosmopolitan subject set adrift in cyberspace, or any space. Bey’s bazaars symbolize the “interzones,” the marginal territories in the networks of intersecting practices of power. These are Woods’s “hidden city within the one now being shaped.” These are Raymond’s free assemblages of the co-developers in the agora. Bey’s TAZs do not exist in either cyberspace or real space but both. He even destabilizes the discursive configurations of cyberspace by playfully proposing new meanings for the now ubiquitous terms “Web” and “Net.” For Bey, the Net is “the totality of all information and communication transfer,” a definition that removes the hermetic place in technoculture for the primacy of the Internet. The Web becomes a counter-hegemonic term:
Generally we’ll use the term Web to refer to the alternate horizontal open structure of info-exchange, the non-hierarchic network, and reserve the term counter-Net to indicate clandestine illegal and rebellious use of the Web, including actual data-piracy and other forms of leeching off the Net itself.
The cathedral embodies the hierarchic structure of information exchange in the Net, while the bazaar is an apt metaphor for the Web, a non-hierarchic network. The counter-Net describes the practices of warez d00dz and pirates. In this mapping of terminology, the cathedral and the bazaar transcend their place as metaphors for software development and become metaphors of political anarchism. No longer restricted to the assemblage of programmers, they now serve as metaphors for the assemblage and communication of all. The Net and the Web both contribute to the TAZ, but the TAZ is an informational potential, something that does not exist in the networks but may be a product of the networks. The TAZ is actually beyond the networks, because it “desires above all to avoid mediation, to experience its existence as immediate.” Similarly, the bazaar, as a metaphor of tactical poststructuralist anarchism, is the model of authenticity, the place where the databanks of postindustrial capitalism can no longer trace your purchases, where technologies of surveillance are not situated, where autonomy is the essence of human communication. The architects of information are operational agents in the disruption of cathedral-building authoritarianism and its hierarchic Net that ensnares us all:
Whether through simple data-piracy, or else by a more complex development of actual rapport with chaos, the Web-hacker, the cybernetician of the TAZ, will find ways to take advantage of perturbations, crashes, and breakdowns in the Net (ways to make information out of “entropy”).
Looking California, Feeling Minnesota
The counter-hegemonic practice of Open Source software development has registered in the realm of praxis as governments around the world embrace Open Source and Free Software in an attempt to “break free of the United States’ lock on the global software market” (Festa).12 Government legislation, first in Brazil, then in other Latin American countries, Europe, and Asia, is attempting to resist the Microsoft monopoly and reduce state expenditures on information technology by forcing government agencies to use Open Source or Free Software “unless proprietary software is the only feasible option” (Festa). Sometimes the rhetoric of this legislation is openly ideological: the city of Florence, Italy passed a motion warning that the protracted use of proprietary software was contributing to “the computer science subjection of the Italian state to Microsoft” (Festa). In a notable decision, the city of Munich recently chose Linux over Microsoft Office for its 14,000 government computers beginning in 2004.13 These countries and cities recognize the potentially significant political, economic, and sociological ramifications of the large-scale adoption of software engineered by the cathedral-building strategies of Closed Source software developers such as Microsoft.14 In the United States in October 2002, Democrat members of the House of Representatives Adam Smith, Ron Kind, and Jim Davis sent a note to 74 Democrats in Congress suggesting that the Linux GNU General Public License (GPL) was a threat to national security. The note asked members to support a letter written by Representatives Tom Davis and Jim Turner asking the head of U.S. cybersecurity policy Richard Clarke to reject “licenses that would prevent or discourage commercial adoption of promising cybersecurity technologies developed through federal R&D” (McMillan). The letter by Smith, Kind and Jim Davis asks that the cybersecurity plan to reject the GPL. Critics suggested Smith’s endorsement might be related to the large political contributions he receives from Microsoft (which resides in his home state). Bradley Kuhn, executive director of the Free Software Foundation, charged, “the rhetoric [of Smith’s letter] is almost word for word what Microsoft has been saying for 19 months” (McMillan).
Despite the encroachment of the discourses of computer science into everyday life, cultural theorists have neglected the very important work of a discourse analysis of the web of texts and practices that shape computer programmers and their code; in particular, the philosophy, history, and psychology of computer programming, represented in texts such as Brooks’s The Mythical Man-Month, Steven Levy’s Hackers: Heroes of the Computer Revolution, and Gerald Weinberg’s The Psychology of Computer Programming for example, offer cultural theorists textual instances for discussing computer programming as a situated knowledge. The first step toward the understanding of the “network society”15 should involve a critical analysis of the code that forms the infrastructure for this society. Law professor Lawrence Lessig posits the following rallying cry in his popular 1999 book, Code and Other Laws of Cyberspace:
We live life in real space, subject to the effects of code. We live ordinary lives, subject to the effects of code. We live social and political lives, subject to the effects of code. Code regulates all these aspects of our lives, more pervasively over time than any other regulator in our life. Should we remain passive about this regulator? Should we let it affect us without doing anything in return?(233)
For Lessig, “open code is a foundation to an open society,” and the architecture of the Internet could make cyberspace more or less regulable (108). “How [the code] changes depends on the code writers. How code writers change it could depend on us” (109). A cultural analysis of the discursive practices that constitute the “code writers,” those I have called “the architects of information,” could contribute to the kind of participatory politics Lessig encourages in this often ignored, but increasingly important, segment of society.
University of Waterloo
Michael Truscello is a doctoral candidate in the Language and Literature program in the Department of English at the University of Waterloo, Ontario, Canada. His dissertation topic is the study of texts and practices that constitute open source software development. He has also published in the field of rhetoric of science, and he co-authored an online course in usability testing for Online-Learning.com. His current research is sponsored by the Social Sciences and Humanities Research Council of Canada.
I would like to thank the anonymous PMC reviewer for suggesting improvements to this paper and especially Andrew McMurry for his insightful remarks on an earlier draft. For CS.
14. Microsoft Senior Vice President Craig Mundie announced Microsoft’s new “Shared Source” policy in what has become know as the May Day Manifesto, a speech given to the New York University Stern School of Business on 3 May 2001. Mundie described Shared Source as the following: “Shared Source is a balanced approach that allows us to share source code with customers and partners while maintaining the intellectual property needed to support a strong software business. Shared Source represents a framework of business value, technical innovation and licensing terms. It covers a spectrum of accessibility that is manifest in the variety of source licensing programs offered by Microsoft.” Other than revealing that Microsoft may be afraid of Open Source’s potential, Shared Source changes very little about Microsoft’s essentially Closed Source practices.
1. On the “inescapably cultural base of technology,” Barbrook and Cameron, in their treatise on the Californian Ideology, make explicit what is often ignored in discussions of the history and philosophy of the Internet revolution, as embodied by texts such as Raymond’s “The Cathedral and the Bazaar”: “Their [the Californian Ideologues’] eclectic blend of conservative economics and hippie libertarianism reflects the history of the West Coast—and not the inevitable future of the rest of the world.” The historicization of the code via the analysis of computer programming history and practices should be a central concern for cultural theorists.
2. Despite some early attempts to theorize the Internet as a space that embodied either surveillance and control or personal autonomy and freedom, the more considered position is that of law professor Lawrence Lessig: “[I argued instead that] the nature of the Net is set in part by its architectures, and [that] the possible architectures of cyberspace are many. The values that these architectures embed are different, and one type of difference is regulability—a difference in the ability to control behavior within a particular cyberspace. Some architectures make behavior more regulable; other architectures make behavior less regulable. These architectures are displacing architectures of liberty” (30).
3. This expression, “the architects of information,” is intended to suggest more than just a resemblance to the information technology sector moniker of “information architect,” a title generously applied to web designers, software engineers, and similar vocations. I use the term “architect” to draw attention to the way software (the code, cyberspace) constructs real space. The term is also used as a metaphor for individual agency, as for instance by David Harvey: “The architect shapes spaces so as to give them social utility as well as human and aesthetic/symbolic meanings. The architect shapes and preserves long-term social memories and strives to give material form to the longings and desires of individuals and collectivities. The architect struggles to open spaces for new possibilities, for future forms of social life. For all of these reasons, as Karatari points out, the ‘will to architecture’ understood as ‘the will to create’ is ‘the foundation of Western thought’” (Harvey 200). Above all, and quite simply, the term “architect” makes explicit (more explicit than, say, “programmer”) the discourse of the codification of space, both real and virtual.
4. As William J. Mitchell says in City of Bits: “In the end, buildings will become computer interfaces and computer interfaces will become buildings” (105).
5. For a recent cultural analysis of the economics of Open Source, see Wershler-Henry. His first chapter contains a brief overview of Mauss, Bataille, Derrida, and the notion of gift economies.
6. While Linux has made serious inroads in the server market, it occupies only 2% of desktops, where Microsoft still dominates.
7. Linux’s success is more evident in the private sector. According to an Information Week survey from March 2002, “more than half of small companies use Linux as an operating system for desktop PCs, as do nearly half of large ones. Nearly 70% of small companies and more than 60% of large ones say they decided to use Linux as an alternative to Windows” (Ricadela).
8. Barbrook and Cameron call Raymond’s brand of technolibertarianism “the Californian Ideology,” a “mix of cybernetics, free market economics, and counter-culture libertarianism that is promulgated by magazines such as Wired and Mondo 2000 as well as the books of Stewart Brand, Kevin Kelly and many others.” Instead of recognizing market forces and techno-cultural discourses as historically situated entities, the Californian Ideology is dogmatic about the inevitable triumph of capital and technology: “Implacable in its certainties, the Californian Ideology offers a fatalistic vision of the natural and inevitable triumph of the hi-tech free market—a vision which is blind to racism, poverty and environmental degradation and which has no time to debate alternatives.”
9. Raymond’s essay is influenced by the techno-libertarian ideas of Michael Rothschild’s Bionomics: Economy as Ecosystem and Kevin Kelly’s Out of Control: The New Biology of Machines, Social Systems, and the Economic World, enormously successful right-wing neo-liberal economic texts. These texts and others trade on the libertarian notion of the markets as “natural” entities that evolved better without government intervention, an amusing contention to be embraced by technology companies on the West Coast since much of the infrastructure that made the dotcom revolution possible was sponsored by the federal government. The most popular distillation of techno-libertarian economics and the naturalistic fallacies they promoted was, ironically, Bill Gates’s Business @ The Speed of Thought: Using a Digital Nervous System. My essay does not have room for a discursive analysis of this genre; however, a continuing study of the historical and ideological roots of the Internet revolution would certainly find these texts of primary importance.
10. This mapping of the transformation of Open Source software development into a political philosophy onto the concepts of poststructuralist tactical anarchism is not intended to be a wholesale exchange of ideas. The intent is, rather, to examine points of generative convergence between the philosophy of software development and certain proponents of poststructuralist thought. The techno-libertarianism from which CatB emerges contains contradictions within it that could never be reconciled with French poststructuralists such as Foucault and Deleuze. As Barbrook and Cameron note of this brand of techno-libertarianism: “the Californian Ideology has emerged from an unexpected collision of right-wing neo-liberalism, counter-culture radicalism and technological determinism—a hybrid ideology with all its ambiguities and contradictions intact.”
11. Lessig argues for the liberatory potential of Open Source from a legalistic perspective, concluding: “Put too simply, everything I have said about the regulability of behavior in cyberspace—or more specifically, about government’s ability to affect regulability in cyberspace—crucially depends on whether the application space of cyberspace is dominated by open code. To the extent that it is, government’s power is decreased; to the extent that it remains dominated by closed code, government’s power is preserved. Open code, in other words, can be a check on state power” (100).
12. It should be noted that perhaps the most important effect Open Source has had on the software industry is less economic than psychological, because for the first time in years a legitimate threat to Microsoft’s dominance seems practical (Wilcox, Gilbert, and Ricciuti).
13. Decisions such as this one may also suggest Open Source is becoming a response to American cultural imperialism.
15. In Manuel Castells’s seminal study The Rise of the Network Society, the logic of “place” is superseded by the ahistorical space of “flows”: “the network of communication is the fundamental spatial configuration: places do not disappear, but their logic and their meaning become absorbed in the network” (443). Castells argues that the “coming of the space of flows is blurring the meaningful relationship between architecture and society” (449). Castells reconceptualizes postmodernism around this point: “In this perspective, postmodernism could be considered the architecture of the space of flows” (449). A discourse of the “architecture of information,” as I have identified it here, might be considered a continuation of this project.