D. Linda Garcia STANDARDS FOR STANDARD SETTING: CONTESTING THE ORGANIZATIONAL FIELD Abstract: Over the past several years, IT standardization has witnessed a breakdown in the consensus that hitherto governed the activities of participants in this domain. Not only have new players appeared on the scene; the legitimacy of the old system has also been called into question. In some cases, the Government has been called on to intervene as a third party. This state of affairs can best be understood as a crisis within an ‘organizational field.’ An organizational field is a recognized domain of human activity in which interdependent actors/organizations interact in accordance with an agreed upon logic or script to achieve some common objective (DiMaggio and Powell 1991). Consensus with respect to objectives does not, however, preclude conflict. To the contrary, organizational fields are highly contested arenas in which actors compete not only for resources but also—and as importantly—for legitimacy (Schwartz 1997). By legitimating their own values and positions within a field, organizations can tilt the ‘rules of the game,’ in their favor with long lasting effects (Fligstein 2002). Viewing a domain of human activities through an organizational field perspective helps to bring the roles of, and the relationships between, public and private realms into greater relief. Generally speaking, organizational fields are relatively self-contained and independent within their own sphere of activity (Schwartz 1997). However, when fields become the battleground of contesting parties, the autonomy of the field is greatly reduced. Contending parties will seek to reinforce their positions by linking them to the deep-seated values and norms of society as a whole (Bourdieu and Wacquant 1992). It is at this point that the public sector has a unique opportunity—or as some might say a major responsibility—to revisit and reevaluate performance within the field in terms of the broader public interest. IT standard setting is an organizational field of activities in which the public and private sectors both have tremendous stakes (OTA 1992). Such is especially the case today given a networked economy in which standards serve not only to determine the architecture and topography of the economic infrastructure, but also winners and losers (OTA 1992; OTA 1994). In a period of rapid technological advance, traditional organizational forms and modes of standard setting are breaking down, while new ones—such as standards consortia, and open source models—are competing intensely for the terrain. There is no reason to believe that the outcome that emerges from this private sector contest will coincide with the public interest (North 1991; Hodgson, Itoh et al. 2001). Thus, as the 1 contenders vie for predominance in the field, the time is ripe for Government, on behalf of the public, to become more engaged in the action by setting some basic standards for standard setters. INTRODUCTION In the United States, much of the discussion about standard setting has focused on the question of ‘what is the appropriate division of labor between the public and private sectors in this arena?’ With few exceptions, the answer has been to assign the responsibility for setting standards to the private sector. 1 This choice has been based on two basic assumptions. First, it is believed that private sector standards are highly attuned to market forces, so they provide the greatest economic value. Second, the government is understood to have no stake in the outcome of standards processes apart from its role as a ‘consumer of standards.’ Thus, by participating in the standards process as a user, the government—it is believed—can fully serve the public’s interest. These assumptions are hardly self-evident. For one, in the case of standards, market forces do not necessarily lead to efficient outcomes. Because standards exhibit many characteristics of public goods, standard setting is subject to considerable market failures (Kindleberger; Farrell and Saloner 1985; Farrell and Saloner 1988; Berg 1989). Moreover, market failures are especially likely in the case of information technology standards, which give rise to externalities and other network effects. 2 Equally problematic—as the 1992 Office of Technology Assessment report to the Congress pointed out—even when voluntary standards organizations step in to facilitate standard setting, bureaucratic failures can delay the process, while conflicts of interest—related in part to standard sales—may serve to distort outcomes (OTA 1992).3 The second assumption—that government has no ‘public interest’ in standard setting—is equally difficult to justify. Standards constitute an infrastructure or platform that supports and sustains the Nation’s economy. As such, standards help to determine the efficiency and effectiveness of the economy, the cost, quality, and availability of products and services, and the state of the Nation’s health, safety, and quality of life. Not surprisingly, therefore, most other governments play a key role in national and international standard setting processes. In fact, given the increasingly competitive global environment, many now link their standardization efforts to their trade policies, employing national standards as marketing devices to attract and lock in customers worldwide.4 (OTA 1992) How, then, might we explain the disparity between our basic assumptions about standard setting and the way the process works in practice? To do so, we must look beyond narrow economic explanations that focus primarily on competitive, market strategies. In addition to market variables, we must also examine cultural and political factors. In particular, we must look at how economic actors compete to construct cultural/cognitive maps that serve not only to reduce uncertainty and generate greater stability among interdependent actors, but also to legitimate and perpetuate the prevailing state of affairs (Fligstein 2002). 2 To capture the entire picture, it is useful to view the standard setting process from an organizational field perspective. In particular, we need to understand how organizational fields serve as economic governance structures. As we shall see, the standards setting process constitutes a unique organizational field that has been governed by a set of legitimating assumptions, many of which are being contested today. ORGANIZATIONAL FIELDS AND GOVERNANCE STRUCTURES Organizations do not exist in a vacuum. They are open systems, subject in isolation to entropy. 5 Organizations are not only enmeshed in a broad range of societal networks; they are also deeply embedded in a shared and all-encompassing institutional environment (Granovetter 1985). While an organization’s networked location helps to determine its resources and power, it is the prevailing set of institutional arrangements—both formal and informal—that provides the incentive structure and the rules of the game by which organizations act (North 1991). To survive organizations must be responsive not only to their own internal dynamics, but to external developments and events as well. To maintain themselves, and to fulfill their missions, organizations must have access both to external material resources as well as to institutional legitimacy. 6 For these purposes organizations must not only respond to their environments; they must actively manage them. As Nohria and Gulati have pointed out, much of an organization’s structure and behavior can be explained in terms of how it is situated to meet these needs. 7 For example, many organizations seek to become chartered by established collective authorities, so they can better legitimize their own goals and structures. Similarly, powerful organizations may seek to embed their goals and procedures directly into society as institutionalized rules. Likewise, organizations may import external resources and elements not because they yield efficiency, but rather because they reinforce the organization’s legitimacy (Meyer and Rowan 1991; Nohria and Gulati 1994). In negotiating their environments to meet their needs, organizations rarely operate on their own. Rather, they typically act in conjunction with other like-minded organizations, which are engaged in similar or complementary activities, so that the success of each is contingent on all others. In any particular market, for example, firms engage routinely with key suppliers, customers, competitors, and regulators. All have an interest in reducing uncertainty and stabilizing their environments 8. In their efforts to do so, these organizations develop, over time, a set of structured practices unique to their institutional space. Once these practices become taken for granted, they serve to recreate and reconstruct the existing economic order of things (White 2001; Fligstein 2002). This set of interdependent actors, as well as the common meaning systems and cognitive maps that govern their interactions, constitute an “organizational field.” As described by Scott, an organizational field is: 3 . . .a population of organizations operating in the same domain, as indicated by the similarity of their services or products. Included also are those others that critically influence their performance, including exchange partners, competitors, funding sources and regulators (Scott 2001). Four interrelated developments are entailed in the structuring of a field. First, there is a growing interaction among organizations. Next, inter-organizational structures of domination and patterns of coalitions begin to appear. Third, the field becomes increasingly complex, having to deal with more and more information. Finally, organizational participants become cognizant of their interdependencies, and aware that they are involved in a common endeavor governed by a shared set of rules (DiMaggio and Powell 1991). Responding to a common environment as well as to each other, organizations within a field eventually become isomorphic9 (Meyer and Rowan 1991). Bounded by shared cultural normative frameworks and a common regulatory system, organizational fields are recognizable as discrete areas of institutional life (Scott 2001). They serve to provide stability for interdependent organizations that depend upon, but also must compete with, one another to create markets and access limited resources. Generally recognized as such, organizational fields enjoy a certain degree of jurisdictional autonomy; in the course of their interactions, they develop rules of the game and ‘conceptions of control’ over a given realm of social, economic, and/or political life (Fligstein 2002). In so doing, they help to reconstruct and legitimate existing social arrangements. As described by Mohr: Organizational forms provide the containers into which the multiple dimensions of institutional life are poured, and organizations combine kinds of persons, types of problems, and forms of treatment in an ongoing and materially recognizable way. Thus it is through the establishment and institutionalization of organizational forms that one or another set of institutional logics comes to be anchored in place (Mohr 2003). As in all governance systems, organizational fields are characterized by contests for power. Fields are organized hierarchically, so that power is asymmetric. It is the most powerful incumbents in the field who define the meaning of the space as well as the roles and relationships of its occupants and the rules of the game. 10 As described by Fligstein: Rules are not created innocently or without taking into account ‘interests.’ If the largest firms are able to work under a set of rules that allows them to dominate the main markets of a society and keep workers disorganized, those rules enforce a system of power (Fligstein 2002). 4 Incumbents can, moreover, strengthen their positions to the extent that they can link their claims to general societal understandings about how things should be done 11. Efficient, socially optimal outcomes in power contests over organizational fields are by no means assured, however. To the contrary, achieving a new consensus is not only a costly process; it is subject to problems of collective action (Olson 1971). These problems will be exacerbated if an organizational field has become so disorganized and unstructured that it has no dominant actor or set of actors to champion a change (Garcia 2003). If, under such circumstances, the uncertainties about the distribution of economic rewards becomes too great, and all stand to lose, the actors in an organizational field may look to the state to broker relationships and/or develop a new consensus and a new set of rules, which can be legitimized at a broader, societal level (North and R.P. 1973; Libecap 1989; North 1991). To appreciate how such a situation might occur, one need only consider the history of railroad regulation. Because of its high fixed costs, fluctuating demand, scale of operations, and need for coordination and specialized engineering skills, the railroad industry was prone to exceptionally high transaction costs. Seeking to achieve greater market stability, the railroad companies alternated between two extreme strategies—cutthroat competition or pooling and price fixing. Neither was successful: cutthroat competition ruined everyone, but cooperative agreements were unenforceable. When private efforts to achieve stability failed, politicians stepped in, establishing the Interstate Commerce Commission in 1887 to regulate this important national industry 12 (Kolko 1970; Kennedy 1991). Organizational fields are thus embedded in and constitutive of the broader institutional environment in which they operate (Granovetter and Swedberg 1992). At the same time, this institutional context serves not only to constrain organizational options; it also helps to determine and reinforce the criteria by which people develop their preferences and evaluate organizational behavior (Berger and Luckmann 1967; Weick 1969; DiMaggio and Powell 1991). Given this mutually reinforcing set of interrelationships, one can understand why new organizational forms are relatively rare (DiMaggio and Powell 1991; Meyer and Rowan 1991; North 1991). Challengers to incumbents of an organizational field must not simply compete for resources based on new forms of practices; as importantly, they must reinterpret current practices in the light of an alternative institutional logic. It is on the basis of this new logic that the flow of resources is then redirected towards the new challengers (Mohr 2003). Not surprisingly, changes in organizational fields are generally associated with basic structural changes at the level of society as a whole. Under radically new circumstances, old organizational arrangements, which serve to buttress a prior and no longer viable set of institutional arrangements, become increasingly problematic (Suchman 1995). Organizational innovations are required not only to redirect attention and behavior towards new ends, but also to reduce uncertainty in a rapidly changing environment (Stinchcomb 1990). 5 STANDARD SETTING AS AN ORGANIZATIONAL FIELD The evolution of standard stetting in the United States parallels that of an organizational field. As described below, as standards began to assume an increasingly important role in the economy, specialized organizations emerged to develop them. These organizations modeled themselves in keeping not only with American social and political traditions, but also with the forms and practices of one another. Their efforts, moreover, were legitimated by the State, which in effect delegated responsibility for setting standards to these private sector organizational entities. This set of arrangements was relatively stable for more than half a century. It is only in the past several years—and in the wake of major technological and socio-economic changes—that newcomers have been able to effectively contest the field. The emergence of standard setting as an organizational field is closely linked to the industrial revolution. With the division of labor and specialization, tasks become more interdependent, requiring greater coordination and information exchange (Durkheim 1933). Standards greatly facilitated these processes. One need only consider the role of standards in mass production, which required interchangeable parts. As described by Harold Williamson: Chief among the other elements in the pattern of mass production is the principle of standardization. Stemming from the rudimentary division of labor, standardization involved the continuous pursuit, and progressive realization, of uniformity of the materials, operations, and products of industry, which made possible the future subdivision and mechanization of labor (Williamson 1951). The relationship between standards and mass production was self-reinforcing. Further advances in precision manufacturing required the development of machine tools and precision gauges, which in turn further drove the need for standards and standardized measures. 13 Standards were also spurred on by the extension of markets across the American continent. As trade became more dispersed, standards were needed to assure that products manufactured in different locals could work together and be easily replicated, assembled, and repaired. Moreover, standards were required to facilitate trading itself. For example, when the railroads extended trade over vast regions, standardized procedures for billing and exchange—such as bills of lading—were required. Likewise, standardized business practices and procedures, as well as standardized time, were needed to coordinate the increasingly complex railroad operations (Kirkland 1961). As the role of standards increased, so did the number of people who had a stake in the standards process. Producers, for example, employed standards as trademarks to differentiate their products from their competitors and to price products for different markets. For suppliers, standards specifications meant reduced production costs. Consumers likewise benefited from standards. They not only conveyed product information and provided greater quality control; standardized products were also cheaper. The general 6 public also called for the development of standards to protect against the growing technological mishaps that were associated with industrialization. (OTA 1992). Although the Federal Government became involved in standards as early as the mid-eighties through the work of the Office of Weights and Measures, and later with the establishment of the Bureau of Standards, it was not until World War I that the government’s stake in standards was really brought home to the Nation. In 1917, product diversity was so great it threatened the war effort. To deal with the problem, the government set up a Commercial Economy Board of the Council of National Defense, whose task was to simplify the use of labor, capital, and equipment for all industries 14 (Cochrane 1966). The government continued its ‘standards crusade’ following the war in and effort to revive the economy. Led by Secretary of Commerce Herbert Hoover, the campaign called for standardization of business practices, materials, machinery and products; specifications to insure high quality products; and product simplification. To promote standardization, the government organized agencies within the Department of Commerce to provide standards assistance to businesses at their request (OTA 1992). As more and more stakeholders became involved in standards, it became increasingly necessary to specialize their operations, and differentiate among these groups. relationship between the public and private sectors. Of prime importance was the Although the government actively promoted standardization at the turn of the century, it gradually relinquished this responsibility to the private standards development organizations. The American preference for private, pluralist solutions is as old as the Constitution itself. From the outset of the new republic, Americans proved to have a penchant for joining factions and establishing associations, a trait that did not escape the observations of Alexis de Tocqueville when he visited America in the mid-1800s. As he described in Democracy and America, “Whenever at the head of some new undertaking you see the Government of France, or a man of rank in England, in the United States you will be sure to find an association.” (deTocqueville 1963). This support for voluntary, private associations was reinforced by a general suspicion of the state and preferences for market-based solutions (Wuthnow 1991). Thus, whereas in many other countries government actively sponsored the growth and development of business, in the United States industrial development was managed, directed, and financed primarily by the private sector (Vogel 1987). The first standards organizations were in keeping with this tradition. Emerging to deal with specific needs as they arose, they took a variety of forms. Often established on an industry-by-industry basis, there was at first little interaction among them. The first American standards organization was the United States Pharmacopial Convention set up in 1829 to establish uniform standards for drugs. The American Iron and Steel Institute, established in 1855, was the first trade association to develop standards. And the American Society of Civil Engineers, which was formed in 1852, was the first scientific and technological society involved in standards development (OTA 1992). 7 With the proliferation of standards organizations, the need for rationalization and coordination of standards activity soon became apparent. Standards organizations were not only competing with one another to write standards, they were also writing conflicting standards, thus defeating the purpose. Coordination among them proved difficult, however, eventually requiring the prodding and promotion of the Federal Government. The first steps at coordination took place in 1918, during the First World War, when five national engineering societies, together with the US Departments of War, Navy, and Commerce, formed the nucleus of an organization that was to become the American Standards Association (ASA). The scope of the ASA’s activities was broadened further during the Second World War, and its constitution was revised to assure that all of those interested in a particular standard would have a voice in its development. The revised constitution required, moreover, that three members at large be included on the association’s board of directors in order to assure a voice for consumer interests (OTA 1992). The broadening of ASAs mandate had only a marginal effect on its ability to serve as coordinator of private sector standards. Many private sector bodies resisted its mandate, not wanting to sacrifice their own independence. To broaden its support the ASA adopted a new constitution and bylaws, and renamed itself the United States of America Standards Institute (USASI). Characterizing itself as a federation of trade and other organizations, the Institute no longer proposed to develop standards, but only to orchestrate their development through the combined technical talent and expertise of its member bodies. As an umbrella, clearing-house organization, it intended to certify that these standards development bodies adhered to the consensus process. The Government and other members of the standards community resisted the effort of ASA to strengthen its role. The Federal Trade Commission protested the use of the name USASI on the ground that it suggested that ASA was an official organization of the Federal Government. A compromise was eventually reached and ASA became the American National Standards Institute. Having no official charter, ANSI became in effect the ‘self-designated’ national coordinating body for US standards development organizations as well as the internationally accepted member body in the International Organization for Standardization (ISO) and the International Electrotechnical Commission (OTA 1992). Although nongovernmental standards organizations in the United States continue to operate in a pluralistic fashion, with only 25 percent coordinating their activities through ANSI, together they constitute an easily recognizable and relatively coherent organizational field. Membership in the standards community has been relatively stable over the years, and the players are well known to one another. Moreover, most of the approximately 400 voluntary standards bodies--including trade associations, professional societies, general membership organizations, and third party certifiers—resemble one another in significant ways. In particular, these organizations all arrive at decisions through a process of consensus, and provide some level of due process. In addition, they all have mechanisms for participation, comment and appeal. Equally—if not more—important, almost all standards organizations are adamant proponents of the voluntary standards process; whatever the disagreements among themselves, they have 8 consistently joined together to defend against any government encroachment on the autonomy of their field. And, by appealing to the American preference for private sector solutions, US voluntary standards organizations have been highly successful in legitimating their right to govern the field. 15 The Federal Government has rarely intervened in private sector standards activities; instead, it has focused on preventing anti-competitive outcomes as well as on assuring the ‘fairness’ of the system. The government’s preference for voluntary consensus standards was reaffirmed in the 1979 Trade Act, which formally recognizes the private sector’s role in standards development, and in the Office of Management and Budget Circular A-119, which directs Federal agencies to use voluntary standards wherever possible in both regulatory and procurement activities (OTA 1992) CONTESTING THE FIELD While staunch in defense of their autonomy, voluntary standards organizations have more often than not been at odds with one another. One major source of their contention has been competition over standard sales. Many of these organizations resemble publishers insofar as they orchestrate standard setting in exchange for the right to sell standards. Dependent upon standard sales as a means of support, competition among standard organizations has often been rife. In this context, members of the standards community typically disagree about which organizations are the most legitimate, as well as about which produce the ‘best’ standards. For instance, many professional societies claim that their standards are technologically superior, since their members participate not as representatives of any group or interest, but rather as individual engineers. On the other hand, many industry-based groups argue the opposite; contending that standards set by professional societies do not reflect market forces, and are thus not adequately sensitive to industry competitive issues. In a similar vein, the American Society for Testing Materials (ASTM) insists that true consensus requires the participation of all interested parties, even if this requires subsidizing some groups. In contrast, the American National Standards Institute (ANSI), as well as others, argue that due process requires only that all have an opportunity to participate. Not surprisingly, given the self-contained nature of the standards community, this type of rivalry has been compounded by personality conflicts, some of which date back several years (OTA 1992). It is important to note, however, that no matter how contentious these conflicts, they have always been self-contained, initiated by incumbents and taking place internally within the standards setting field itself. Moreover, rarely—if ever—did these incumbents question the overall functioning or legitimacy of the system. Thus, it is only recently that the standards setting field itself has become the subject of debate. As one might anticipate, the major challengers are newcomers to the field, having emerged to address the standards issues associated with advances in information technology and the advent of the networked economy. 9 The present challenge stems from the fact that the formal, voluntary consensus-based standard process is not well equipped to deal with rapidly advancing information-based networking technologies. Relying on the slow and often arduous process of consensus building, standards bodies are hard pressed to keep pace with changing technologies. In an effort to make allowances for technology change, and facilitate interoperability among an increasing number of interdependent parties, networking standards have often been incorporated into elaborate reference models and defined in very broad and generic terms. These types of standards are typically referred to as “anticipatory standards,” because the process of setting the standard anticipates the creation of the product. The problem with these standards, however, is that, even after such standards have been formally set, users still have to implement compatible technologies that meet standards specifications; and products need to be certified as to their compatibility with one another (Cargill 1989). This process can be so complex and time-consuming that the window of opportunity sometimes closes and those standards are overtaken by new technologies and events. In the late 1980s, many vendors, discouraged by this lagging process, started to circumvent the traditional standards-setting process and to develop standards in more focused interest-based consortia. Included among these, for example, were consortia to develop standards for Switched Multimegabit Data Service (SMDS), Fiber Distributed Data Interface (FDDI) over twisted pair, asynchronous transfer mode (ATM), and frame relay technologies. The major user consortia included the Corporation for Open Systems (COS), Manufacturing Automation Protocol (MAP), and the Technical Office Protocol (TOP) (Weiss 1995)and Cargill 1995). Operating in a relatively closed environment, these groups greatly simplified the standards process. Unlike traditional standards organizations, consortia are not bound by rules guaranteeing openness and consensus. Membership is generally restricted to those willing to pay a significant fee. Because of this exclusivity, these consortia often replicate the dynamics of the market; participants invest in the process because of an anticipated ‘payoff.’ One consortium that stands out for its success in achieving both openness and speed is the Internet Engineering Task force (IETF), responsible for developing standards for the Internet. The IETF’s open process owes much to the Internet’s unique history. Like the network itself, Internet standards evolved in a very informal way as part of the efforts of the Defense Advanced Research Projects Agency (DARPA) to establish computer networks linking researchers across the country. Because the participants were few, and bound together by a common research purpose, they were able to develop standards in a relatively timely fashion. But, even though the Internet has grown by leaps and bounds, the IETF has held to its tradition of openness and inclusivity. By employing the Internet to conduct on-line forums and provide access to standards and standards related, the IETF has greatly reduced the financial barriers to participation. Equally important, the IETF process avoids the implementation and conformance-testing problem associated with anticipatory standards because all specifications need to be implemented and demonstrated to be interoperable. Similarly, to become a full standard, a draft standard must be fieldtested and proven capable of maintaining a community of Interest over time. 10 Spurred on by the Internet and the growth of the Word Wide Web, standards setting proliferated in the 1990s, leading to a major shift in the standard setting field. The World Wide Web Consortium (W3C) was established in 1994 to meet the demand. Designed along the lines of the IETF, but on a more formalized basis, the W3C provided a successful model that others soon emulated. Eager to become part of the action, technology manufacturers and vendors bypassed traditional standards organization and set up consortia to develop standards that were compatible with the Internet and the World Wide Web. Those that did so were highly successful; whereas the standards developed through the more traditional organizations—such as ANSI’s X.12—failed to take hold. Today, with more than 260 consortia, it is clear that consortia have become the organization of choice for IT standards developers (Spring and Cargill) Cargill 2002b. As described by one observer, ‘a new industry consortium is founded ever week.’ At the same time, a number of intermediary groups has emerged to support them by registering and cataloging their standards (Libicki). Based on an entirely new business model, the Open Source movement represents a much more radical shift away from the traditional mode of standard setting. As one might suspect, participants in the movement are all newcomers to the standards field. Like the internet, the open source movement emerged in the context of a distinct culture of openness and reciprocity dating back to the late 1960s and 1870s when programmers freely shared their source code. In fact, the movement was born in direct response to the renunciation of this norm by computer companies that decided to proprietize their operating system software (Weber 2002). Richard Stallman, a programmer at the MIT Artificial Intelligence Laboratory, led the charge establishing the Free Software Foundation in 1984 with the goal of generating hacker culture by creating a pool of free software, utilities and programming tools (Glyn 2001). To support this effort, Stallman codified these norms in a unique form of property rights—the General Purpose License (GPL), or copyleft as it is also called. Under this arrangement, software programmers were permitted to modify and enhance open source code, but only if it remained open and was provided to others on the same basis. Although Stallman’s project rallied the faithful and provided a vision for the open source movement, it was unable to create a critical mass of supporters, due in part tot its strong ideological tone (Moody 2001). In 1991, Linus Torvalds, a 21-year-old computer science major at the University of Helsinki, had much greater success. Building on a simplified version of Unix, Torvalds created Linux, the kernel of a new operating system. Posting it to the internet, he invited his peers to modify and enhance it on the condition that functions written by others could also be freely distributed. Torvalds tapped a large, eager audience and within a year 100 people had joined the Linus newsgroup. Ninety renditions and 2 ½ years later, Linux was released to the public (Moody 2001). Today, the open source movement has a considerable following, and some open source software products are leaders in their application spaces. As of November 2001, for example, the web server Apache was used on approximately 62 percent of active servers across all domains, while Sendmail was used for 75 percent of internet e-mail. Linus now has 21 million users worldwide. Impressed by this growth, many 11 major corporations, including Hewlett Packard, IBM, and Dell, are developing and using open source software. Even more telling, Microsoft has identified Linus as its chief competitor both in its international communications as well as in antitrust proceedings. The big question is whether or not the open source movement can maintain this momentum without generating private claims on residual resources. For the present, at least, private claims are not a problem. The movement’s institutional structure, together with its culture of reciprocity and sharing has served to maintain the common pool problems typically associated with open access16 (Frank and C. June 2002). Faced with such competition, traditional standards development organizations have not remained on the sidelines. Recognizing that information technology standards represent a new and unique challenge with respect to both time and the need for interoperability, these organizations sought to streamline their procedures and speed up their processes. ANSI, for example, set up a fast track program for IT standards, whereby ANSI provides accreditation for standards that had been developed in other, consortia-like forums. Relatively few groups took advantage of it, however. Reforms were also undertaken at the international level in organizations such as ISO, which adopted a Publicly Available Specification Process, allowing standards adopted in other forums to move quickly through the ISO acceptance process. Again, notwithstanding these changes, most companies continued to place their bets—as well as their financial resources—on consortia. To maintain their privileged positions, traditional standards bodies continually seek to define the standards field in their image. In particular, in promoting their cause, they have emphasized the open, voluntary consensus aspect of their organizations, linking it to the American political tradition of openness and voluntarism. To this end, for example, these organizations have contended that the US law mandating the government’s use of voluntary consensus standards excludes—by definition—standards developed within consortia. As described, for example, in ANSI’s strategic plan: [the U.S. government should encourage more use of the principles embodied in accreditation by recognizing the ANSI process as providing sufficient evidence that American National Standards (ANS) meet federal criteria for voluntary consensus standards. [In addition] Non-traditional standards organizations should review their objectives to determine whether closer interaction with the formal system will help add value to their efforts (as cited in (Cargill 2002b)) Faced with this opposition, the consortia have been no less adamant in touting their claims. As described by Cargill, “. . .both groups believed they were in a ‘do or die’ situation, in which their specific survival depended upon their ability to diminish (but not destroy) the other side whilst retaining hegemony for their set of beliefs (Cargill 2002a)). In legitimating their approach, the consortia have focused on economic issues, while downplaying those related to process. Accordingly, they characterize their policy of restricting access to those who ‘pay to play,’ as the key to their success. What matters, they 12 contend, is not that the process is open to all, but rather that standards are developed, which are widely used. As proof, they point to the willingness of so many businesses to invest their funds in consortia. This emphasis in their rhetoric on economic outcomes is not surprising, given that the rational and legal basis for setting up consortia is the National Cooperative Research and Production Act of 1993, the purpose of which was to support the creation of business alliances as a means of promoting US competitiveness. Joining together in what might be characterized as a cause-based social movement, advocates of open source software have likewise sought to redefine the field of standard setting. And much like the proponents of competing approaches, they use rhetoric that links their way of doing things to fundamental social values. Thus, for example, to overcome any negative connotations associated with the original name “free software,” the movement’s leaders substituted the name of “open source software.” Similarly, open source advocates characterize their movement as the Magna Carta for software users on the grounds that it provides users the freedom to choose. Their argument has resonated with many who, when faced with an economic downturn and suspicious of dominant companies such as Microsoft, have welcomed such lower cost non-proprietary solutions. Opponents of open source software have responded in kind, questioning the fundamental legitimacy of the open source process as well as the quality of the product. Among the movement’s most vociferous opponents is, of course, Microsoft. After originally downplaying the importance of open source software, Microsoft began a major begun public relations and lobbying campaign to undermine its credibility. To this end, it has created a counter-narrative, accusing the open source movement of being both anti-capitalist and un-American (Shanland & Wilcox 2001; (Kerstetter, Hamm et al. 2003). Microsoft has also challenged the viability of the process, arguing that the GPL is unenforceable in court. At the same time, it has actively lobbied the government against the use of Linux on security grounds. Paradoxically, as the role of IT standards has loomed larger, the field of standards setting has experienced greater disarray. Given much higher stakes, companies are competing more intensely than ever to define standards, forum shopping whenever the need arises. The ‘pay to play’ basis of many of today’s consortia facilitates this process; denied the prize in one standard’s arena companies have little compunction to seek out or establish another, where they may have more to gain. Under the circumstances, organizational turnover is relatively high, so the institutionalization of the process is low there is no general, or transparent set, of principles that govern these standards activities. At the same time, the traditionally more open and well established IETF is becoming bogged down not only by the demand for standards but also by the its commitment to achieve a rough consensus among participants. As a result, there has been a multiplication of working groups, and a splintering off of discontents (Libicki ). Traditional standard setting organizations are similarly in a state of flux. All in all, the situation is not a promising one. Given the lack of cohesion within the standards field, it will be difficult to generate number of standards and the level of interoperability required for today’s networked society. . 13 STANDARDS FOR STANDARD SETTERS Actors in an organizational field enjoy a certain measure of autonomy by virtue of the fact that their activities are generally consistent, or at least congruent with, the needs and traditions of the larger environment of which it is a part. When there is considerable dissension within the field, and the gap between private and public goals becomes very great, those field participants who are most discontented will likely bring their issues to the policy realm. And, if the stakes are high enough, the government may decide to intervene. (Garcia 2003). Such is the case in the standards field today. Paradoxically, it appears that, precisely because the stakes are so high, the ability of standards organizations to set standards is inversely proportional to the need for them. As evidenced by the growth of competing standards consortia, businesses are more inclined than ever before to employ standards processes strategically, so as to gain competitive advantage. An open, interoperable architecture is not like to be the result. Yet, as the advent of the Internet and the World Wide Web made clear, to reap the full economic benefits of ICTs, networks and network components must be interoperable and open for interconnection. Open standards reduce transactions costs, as well as the costs of barriers to entry, thereby encouraging competition. Moreover, interoperable systems reduce networking costs, thus encouraging technology diffusion. In addition, open standards encourage innovation by providing a platform on which businesses can add value and develop new products and services. What might be the Government’s response? Historically, as we have seen, the government has intervened in the field of standards setting at those points in time when the nation’s security and economic welfare were at stake. However, the role that the government played was for the most part indirect; instead of setting standards, the government helped to facilitate and promote the standards setting process by playing an informational and brokerage role. In addition, the Government established the normative and legal framework within which standards organizations operate. Setting standards for standard setters is precisely what is needed today. The standards perspective that served well for an industrial age, in which the primary need was for product standards, is no longer appropriate in a networked society, in which the overriding requirement is for interoperable standards that can provide an open platform to support all economic, political, and cultural activities. Unfortunately, in today’s hyper ‘deregulatory’ environment, there is no voice, such at that of former Secretary of Commerce Hoover, to articulate a coherent vision of the role of standards and standards organizations and how—in a networked society— they relate to our fundamental values such as first amendment rights, national security, intellectual property, anti trust, etc. 14 Berg, S. (1989). "Technical Standards as Public Goods: Demand Incentives for Cooperative Behavior." Public Finance Quarterly 17: 3553. Berger, P. and T. Luckmann (1967). The Social Construction of Reality. New York, Doubleday. Bourdieu, P. and L. J. D. Wacquant (1992). An Invitation to Reflexive Sociology. Chicago, IL, The University of Chicago Press. Cargill, C. (1989). Information Technology Standardization: Theory, Process and Organization. Boston, MA, Digital Press. Cargill, C. (2002a). Uncommonly Commonality. The Standards Edge. S. Bolin. Ann Arbor, MI, Bolin Communications. Cargill, C. (2002b). The Role of Consortia Standards in Federal Procurement. The Standards Edge. S. Bolin. Ann Arbor, MI, Bolin Communications. Cochrane, R. C. (1966). Measures for Progress: A History of the National Bureau of Standards. Washington D.C., National Bureau of Standards. deTocqueville, A. (1963). Democracy in America. DiMaggio, P. J. and W. W. Powell (1991). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. The New Institutionalism in Organizational Analysis. P. J. DiMaggio and W. W. Powell. Chicago, IL, Chicago University Press. Durkheim, E. (1933). The Division of Labor in Society. New York, NY, Free Press. Farrell, J. and G. Saloner (1985). "Standardization, Compatibility, and Innovation." Rand Journal of Economics 16(spring): 70-83. Farrell, J. and G. Saloner (1988). "Coordination Through Committees and Markets." Rand Journal of Economics 18(summer): 235-252. Fligstein, N. (2002). The Architecture of Markets. Princeton, NJ, Princeton University Press. Frank, E. and J. C. (June 2002). "Reconciling Investors & Donators: The Governance Structure of Open Source,". Zurich, University of Zurich. Garcia, D. L. (2003). Glyn, M. (2001). Rebel Code: Inside Linux and the Open Source Revolution. Cambridge, Massachusetts, Perseus Publishing. Granovetter, M. (1985). "Economic Action and Social Ctructure: The Problem of Embeddedness." American Journal of Sociology 91(3): 481-510. Granovetter, M. and Swedberg (1992). Hodgson, G. M., M. Itoh, and, et al. (2001). "Introduction,". Capitalism in Evolution: Global Contentions--East and West. M. I. Geoffrey M. Hodgson, and Yomokawa Nobuharu. Cheltenham, UK, Edward Elgar. Kennedy (1991). Kerstetter, J., S. Hamm, et al. (2003). The Linux Uprising. Business Week. Kindleberger. Kirkland, E. (1961). Industry Comes of Age: Business, Labor, and Public Policy. New York, New York, Holt, Rheinhard, and Winston. Kolko (1970). Libecap, G. D. (1989). Contracting for Property Rights. Cambridge UK, Cambridge University Press. Libicki. Meyer and Rowan (1991). Mohr, J. (2003). Nohria and Gulati (1994). North, D. (1991). Institutions, Institutional Change and Economic Performance. North, D. and T. R.P. (1973). The Rise of the Western World: A New Economic History. New York, Cambridge University Press. Olson, M. (1971). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, Massachusetts, Harvard University Press. OTA (1992). Global Standards: Building Blocks for the Future. Washington DC, US Congress, Office of Technology Assessment. OTA (1994). The Electronic Enterprise: Looking to the Future. Washington DC, US Congress, Office of Technology Assessment. Schwartz, D. (1997). Culture and Power: The Sociology of Pierre Bourdieu. Chicago, Il, University of Illinois Press. Scott, W. R. (2001). Institutions and Organizations. Thousand Oaks, CA, Sage Publications, Inc. Spring and Cargill. Stinchcomb, A. L. (1990). "Social Structure and Organizations". Handbook of Organizations. J. G. March. Chicago, Illinois, Rand McNally. Suchman (1995). Vogel, D. (1987). Government-Industry Relations in the United States: An Overview. Comparative Government Industry Relations. S. Wilks and M. Wright. Ox ford, UK, Clarendon Press. Weber (2002). "The Political Economy of Open Source,". BRIE Working Paper 140, E-conomy Project. Weick (1969). Weiss, S. a. (1995). White, H. (2001). Markets from Networks: Socioeconomic Models of Production. Princeton, NJ, Princeton University Press. Williamson, H., ed. (1951). The Growth of the American Economy. New York, NY, Prentice Hall. Wuthnow, R., ed. (1991). The Voluntary Sector in Comparative Perspective. Princeton, NJ, Princeton University Press. 15 The author would like to express her gratitude to Becky Lentz, Program Officer, the Media, Arts and Knowledge Unit, and the Ford Foundation, for funding her research on policymaking and organizational fields. She would also like to thank and acknowledge the help and intellectual input of Ellen Surles, Al de Plazaola, and Kathryn Cornelius— the three Georgetown University, Communication, Culture and Technology Masters candidates collaborating on this Ford Grant. 1 This preference for the private sector was reaffirmed in the 1970 Trade Act, which formally recognizes the private sector’s role in standards development, and in the Office of Management and Budget (OMB Circular A-119, which directs Federal agencies to use voluntary standards wherever possible in both regulatory and procurement activities. 2. For example, network technologies often have large installed bases, making it particularly costly for users to shift to a new, more technologically advanced standard. Thus users may fail to adopt a superior standard due to what economists call ‘excess inertia.’ At the same time, these technologies also exhibit ‘increasing returns to adoption,’ a situation that occurs when the benefits to the user of a technology increases with the number of users. Under these circumstances, the wrong standard might be chose due to ‘excess momentum.’ Not wanting to be left off the network when a major user moves to a new standard, other users may rush too quickly to jump on the bandwagon. 3 As the OTA report pointed out, for example, “The interests of some standards organizations are beginning to diverge from those of manufacturers. In a highly competitive global economy, for example, it is important for manufacturers to have their standards adopted on an international basis. They may even want to ‘give’ their standards away in an effort to develop new markets. However, such a policy is not in the interest of those standard setting organizations, whose livelihoods generally depend on standard sales” (OTA 1992). 4 More and more industries are not only dependent upon trade but also affected by standards. It was estimated, for example, that for the year 1977, $69 billion of US exports were affected by standards activity. No comparable figure is available today. However, in 1992, it was estimated that of $85 billion in exports of manufactured goods, some $45 billion was subject to European Community product safety standards alone. 5 As described by Katz and Kahn, “To survive, open systems must move to arrest the entrophic process; they must acquire negative entropy. The entropic process is a universal law of nature in which all forms of organization move toward disintegration or death. Complex physical and biological organisms also run down and perish. The open system, however, by importing more energy from its environment than it expends, can store energy and can acquire negative entropy. There is then a general trend in an open system to maximize its ration of imported to expanded energy, to survive even during periods of crisis to live on borrowed time (Katz and Kahn ). 6 Legitimacy can be defined as ‘a generalized perception or assumption that the actors of an entity are desirable, proper, appropriate within some socially constructed system of norms, values, beliefs and definitions” (Suchman 1995: 574). 16 As described by Nohria and Gulati, “. . .a firm’s structure is influenced on the one hand by the institutionally 7 legitimate solutions for a broadly defined task environment and on the other by the pragmatic concerns of addressing local contingencies. This tension is mediated by the rhetoric that managers employ and the actions that they take to create, present, and account for the organization’s structure” (Nohria and Gulati 1994: 530). 8 As described by Fligstein: “There are four threats to a firm’s survival. First, suppliers can control inputs, raise prices, and make firms who require their inputs unprofitable. Second, competitors can engage in price competition, take over market share, and eventually drive the firm out of business. Third, gaining cooperation from managers and workers in the firm present problems of interpersonal conflict and politics that can jeopardize the ability to produce goods and services as well. Finally, products may become obsolete. These problems are most acute under conditions of economic turbulence that occur most frequently at the beginning of a market, but that also can reflect a sudden downturn in the market (Fligstein 2001: 17). 9 DiMaggio and Powell identify three forces leading to isomorphism: formal and informal pressures exerted by other organizations; a common legal environment, and organizational benchmarking, especially during periods of uncertainty (DiMaggio and Powell (1991). 10 As Fligstein points out: “Markets (and this includes almost all modern production markets) are mainly structured by sellers looking for buyers. A given market becomes a ‘stable market’ (i.e., a field) when the product being exchange has legitimacy with customers, and the suppliers of the good or service are able to produce a status hierarchy in which the largest suppliers dominate the market and are able to reproduce organizations to make the good and create social relations between competitors to govern competition (Fligstein 2001:31). 11 Characterizing Bordieu’s perspective on this subject, Swartz notes: Both the dominant establishment and the subordinate challengers, both orthodox and heterodox views, share a tacit acceptance that the field of struggle is worth pursuing in the first place. . . .Challengers and incumbents share a common interest in preserving the field itself, even if they are sharply divided on how it is to be controlled—referred to as the doxa. Entry into the field requires the tacit acceptance of the rules of the game, meaning that specific forms of struggle are legitimated whereas others are excluded (Schwartz, 125). 12 The rail owners were not alone in questioning whether the market, functioning on its own, could solve the problem of too much competition. Because the railroads were at the center of national activity, the bankers were among the first to intervene. To force a solution, J.P. Morgan—acting as a neutral third party—sequestered many of the key railroad owners on his yacht. Acting under duress, they were able to come to terms. However, they were quick to renege on their agreement when there was no longer a means of enforcing it (Neuman, et al, 1997). 13 Of particular importance, for example, was the vernier caliper, which was first made in the United States in 1851. Inexpensive and capable of reading to thousandths of an inch, the new caliper permitted ordinary machinists— whether they were gun smiths, watchmakers, or sewing machine manufacturers—to develop precision, interoperable parts (Green 1951) 17 14 According to Cochrane, “Labor savings in the manufacture of products from clothing to coffins reportedly reached as high as 35 percent. Savings over prewar consumption of materials in some instances rose to 50 percent as simplicity ruled and plentiful wood, paper, zinc, and cotton replaced the steel, tinplate, copper, brass, bronze, pig tin, nickel, and raw wool consumed by the war. The country had experienced nothing like it before.” (Cochrane 1966: 167). 15 The division of labor between the public and private sectors has strong support in the US standards setting community. At hearings held in 1990 by the National Institute of Standards and Technology to determine whether the government should be more active in standards setting, especially in the international arena, the response if those testifying was an emphatic “No.” See proceedings, National Institute for Standards and Technology Public Hearings, “Improving U.S. Participation in International Standards Activities,” April 3, 1990. 16 Free riding, for example, does not have a negative impact in the context of open source. Contributing their time and expertise in part to secure their reputations, open source programmers benefit from an extended audience. Because leadership positions are allocated in accordance with a peer reviewed assessment, rewards adequately reflect a programmer’s contribution. Equally important is the General Purpose License, which generates trust by preventing defection. Other things being equal, programmers will continue to contribute to the movement so long as they believe no one can privatize their contributions (Franck and Jungwirth 2002). 18