Universitites, Information Ownership, and Knowledge Communities

The recent decision by MIT to post the information from all its 2,000 courses free to the Web has generated tremendous excitement online, with more than 42 million hits recorded in the first month, according to MIT statistics  [1]. The project, entitled OpenCourseWare, was initiated by MIT professors and funded by $11 million in grants from two foundations. As of March, 2004, 700 courses, encompassing all five schools and two-thirds of the faculty on the Cambridge, Massachusetts campus, have been added to the site (ocw.mit.edu).

The project did not start as an effort to populate the information commons. On the contrary, in 1999, Robert Brown, MIT's provost, asked a faculty committee to study the idea for an online for-profit equivalent to the physical school. However, after researching the issue, the faculty committee concluded that a profit-making venture was not viable, suggesting instead that the university and its faculty make its course material available for free online  [2].

As reported by Charles Vest  [2], the university's president, the OpenCourseWare initiative has had impacts both inside and outside the university. Within MIT, professors have begun using one another's materials to supplement their own teaching efforts, and are discovering interdisciplinary connections that could lead to new innovations inside the institution. Outside the university, MIT alumni, interested individuals, and other educators from around the world are using the courseware as a means to keep current in their fields and as models for new courses and curriculum.

The effort has generated interest in other areas, particularly among Intellectual Property legal commentators, who questioned the relationship between faculty-generated course notes and university property rights  [3]. Given the fact that the project is faculty-initiated and voluntary, intellectual property issues in the curricular area between the university and professors have not yet come to a head at MIT. However, the project has had to navigate the murky waters of copyright in other respects, particularly with regard to the negotiation for permissions with other information providers  [4].

Nevertheless, the project still leaves open the question of the relative information rights of professors and universities. In addition, it raises broader questions of the roles both of professional disciplines and the institutional structures developed to support them in a technological world in which traditional boundaries between information transformation, production, and dissemination are under strain. The following attempts to lay out some of the relevant issues, focusing particularly on the role of the university in an online world.

A Brief Look at the University in Society

Lying at the center of questions about university and academic information ownership is a deeply contested vision of the role of both scholarship and the institutions designed to support research. Do scholars labor primarily as individual authors and inventors, or are they members of what Enlightenment scholars termed a res publica, loosely defined as a republic of ideas operating beyond institutional and political boundaries? Are universities places of sanctuary for ideas, separated from the marketplace, or information dissemination institutions situated squarely in the market?

In her book "Who Owns Academic Work?," Corynne McSherry  [5] traces the history of modern American universities and makes a strong case that these questions are largely unanswerable, because they assume a stability in self-conception that is historically missing. She argues that medieval universities and guilds were primarily envisioned as mechanisms for monopoly control over ideas, with the former focusing on professional control and the latter on control over invention. With the coming of the Enlightenment, voluntary academic societies sought to break down university monopolies on knowledge, constructing a meritocracy based on open communication and communal enquiry, and existing in cooperation with the growing commercial marketplace. At the institutional level, nineteenth-century German conceptions of the university, based on Kant's ideas in Conflict of the Faculties, envisioned the university as a place apart from the marketplace, yet poised to provide knowledge based on reason to political rulers. In the United States, German models of scholarly independence blended with the British tradition of liberal arts and informed citizenship, leading to a tension between disinterested scholarship and community. This admixture was further complicated by the presence of private schools funded through religious and other associations sitting cheek-and-jowl to land-grant public universities, developed to provide practical assistance in the development of new agricultural and mechanical techniques.

By the twentieth century, the split between theoretical and practical knowledge within universities was institutionalized through a separation of faculties of arts and science from engineering and professional school. At the same time, the continued compartmentalization of knowledge into disciplines supported the rise of self-contained academic communities with different standards of scholarship and practice.

To support the engagement of the university in the marketplace, during the 1920's several American universities, particularly those with large engineering components, inaugurated small offices dedicated to technology transfer, particularly the processing of patent applications for professors. However, in a major shift, the end of the Second World War saw a major increase in government grant programs for basic research, insulating the academy from a necessity to rely on private funding sources and enhancing the traditional notion of universities as the preferred site for basic objective research separate from the commercial marketplace. At the same time, a greater integration of the university into public life occurred, with the provision of GI Bill grants to returning members of the military. University enrollments doubled during the next 15 years, doubling again within another 8 years.

By the 1990s, the position of universities within society began to shift again. Federal funding for research slowed, along with other public financing sources. Pressure developed to seek private financing through partnerships with foundations and corporations. Universities undertook attempts at more aggressive management of intellectual assets, often bringing them into conflict with academic communities. The rise of the Internet signaled the potential for developing new resource streams through the development of online courses and degrees, but no one was sure where the dividing line stood between individual and institutional ownership of course materials. Academic publishing, long a backwater in the publishing industry, showed strong growth and consolidation as publishers embraced electronic dissemination and new models of product bundling.

Situating Universities in the Information Policy Universe

Given these trends, the university sits today on two major information policy fault lines, as highlighted in Figure 1. The first (horizontal) line separates transformation activities, including the development of new ideas and texts from prior works, from the process of information production and dissemination. Disputes along this axis occur between stakeholders engaged in the creation of works (such as authors) and those groups like publishers who are primarily concerned with information dissemination.

The second (vertical) line separates assumptions about the obligations of members of a society relative to information use  [6]. On the left side of Figure 1, information policy is seen as supporting communal needs, and is governed by a gift economy  [5]. The right side is based on individualistic notions of information creation and ownership, in which exchange is mediated through markets. Disputes along this boundary often occur when advocates like libraries that support open sharing of information tangle with private owners of information, such as authors and publishers.

FIGURE 1: The University in the Information Policy Environment

As a home for research, the university and its inhabitants are deeply engaged in information transformation. However, the university is also committed to information production and dissemination, both through teaching and community service. Similarly, the institution vacillates between a system that treasures the information commons (seeing its participants as engaging in activity that is both communal and cumulative) and one in which both the institution and its employees act as self-interested individuals in a market economy inhabited by competitors and consumers.

Negotiating this environment requires the use of mechanisms that allow the institution and its inhabitants to bridge seemingly dichotomous realities. As shown in Figure 2, in the case of university this requires the construction of mechanisms for legitimating simultaneous presence in two or more sectors; allowing, for instance, participants to justify their actions based on the norms of both individual and communal transformation. In the information policy world this is achieved primarily through four mechanisms: boundary objects, legal fictions, contracts, and institutional norms.

FIGURE 2: Bridging Information Policy Gaps

Boundary objects are defined as "objects which both inhabit several intersecting social worlds—and satisfy the informational requirements of each of them"  [7, p. 393]. In the case of universities inhabiting multiple information policy realms, boundary objects are terms or symbols that provide a bridge across the fault lines previously discussed. The term 'academic authorship,' for instance, seeks to evoke both the process of academic information transformation in a communal milieu and the state of romantic individualistic authorship critical to intellectual property claims. Similarly, 'self-publishing' connects individual authorship and information dissemination. Historically, 'academic publishing' was designed as a mechanism to assist in the publication of theses by members of academia after the private publishing industry had ceased to provide reasonable alternatives for publication of scholarly monographs [8], thereby allowing the continued free flow of knowledge across academic institutional boundaries. Recently, a traditional norm and two new concepts have developed to bridge the gap between communal transformation and dissemination. Both 'web-based scholarship' and 'open access systems' try to capture the notion of information transformation and sharing in a communal environment not constrained by market priorities. Together, they envision a feedback system where open information sharing leads to rapid transformation and extension of prior research, and subsequent communal sharing of new insights. The norm against plagiarism serves as a boundary object to mediate between the academic requirements for open dissemination and the need to delineate an idea's paternity.

The use of boundary objects is not the only means by which domains may inter-penetrate. Within legal systems, various legal fictions  [9] have evolved to manage the boundaries between domains. As defined in the Encyclopedia Britannica, a legal fiction is "a rule assuming as true something that is clearly false. A fiction is often used to get around the provisions of constitutions and legal codes that legislators are hesitant to change or to encumber with specific limitations."  [10] 'Fair use' and the 'idea/expression dichotomy' can be viewed as attempts to mediate between the demands of proponents of information individualism for control over their creations, with communal requirements to maintain a free flow of ideas in society. In the production & dissemination realms, fictions like 'IP term limitations' and the 'First Sale doctrine' attempt to balance individual ownership of information products with the needs of the community to share knowledge for communal cohesion and sense making. Similarly, the 'work for hire' fiction that treats corporations as individual authors for the purpose of copyright allows the integration of authorship and production ownership for works where there are multiple participants in creation.

It is interesting to note that no legal fiction exists between the communal domains. This can be explained by the fact that legal systems are focused on individual ownership rather than communal norms. As such, they have not adequately explored the legal boundaries between communal sectors  [11]. Boundary spanning in this area is left to less global or formal mechanisms such as individual contracts and citation norms.

Contracts are legal mechanisms used to define the boundaries between information domains. Academic employment contracts, artists' contracts with publishers, researcher contracts with pharmaceutical companies, public and academic licensing agreements with corporations over control of scholarly journal publications, government research contracts with university researchers, and Open Source contracts among LINUX software developers are all examples of attempts to navigate the boundaries between information domains based on individual negotiations.

Finally, institutions seek to manage their balance astride the fault lines through institutional and disciplinary norms. As defined by Axelrod, "A norm exists in a given social setting to the extent that individuals usually act in a certain way and are often punished when seen not to be acting in this way [ [12, p. 96]. According to complexity theory, norms are maintained by punishing people who defect from social rules, punishing individuals who do not enforce rules against defectors, and promoting interaction across group boundaries through the sharing of similar cultural features  [12]. University and disciplinary communities exhibits all these behaviors, through actions such as tenure and peer review processes, university codes of honor, academic and professional review boards, and disciplinary conferences and journals.

Expanding on the earlier notion of universities standing astride two major fault lines, we can envision the competing pressures they are experiencing through an analogy with plate tectonics, whereby movement of the four domains under study sometimes causes a build-up of pressure along boundary lines, eventually precipitating 'continental drift'. Under conditions of pressure, the bridging mechanisms, whether boundary objects, legal fictions, contractual traditions, or institutional norms, come under stress, and are forced to redefine themselves or give way to new mechanisms. Figure 3 highlights some of the forces playing on universities and their faculty. The arrows should be viewed as lines of stress radiating outward from their domain of origin.

FIGURE 3: Information Policy Stress Lines

Certain aspects of the diagram deserve further comment. First, the greatest number of lines of stress radiate outward from the Individual Production and Dissemination domain. This is natural, since the best-resourced economic and social forces in the information arena are located in this space. If universities wished to align themselves tactically with forces radiating the most power within information policy, they would define their mission as an information provider, promoting online courses with strong digital rights management, focusing on research funding from private sources, applying government research patent rules that make them the patent holder of first resort, and re-negotiating their contracts with academics toward a 'work-for-hire' model of authorship. Similarly, academics, if so inclined, would emphasize their status as individual authors and inventors, taking advantage of patent and copyright laws as well as government patenting rules that advantage individual over collective efforts, while using web authoring as a means to promote individual authorship and information dissemination in opposition to the traditions both of communal authorship and corporate information production and dissemination.

Two other aspects of the figure bear discussion. First, the existence of more stress lines indicates the likelihood that the bridging mechanisms in these areas will probably undergo the greatest stress. Secondly, stress is most likely to occur between initiatives operating in opposition to one another. For instance, movements toward individual web authorship would tend to conflict most directly with online communal scholarship and the move by information disseminators to establish online course offerings. Returning to Figure 2, the bridges in these areas are the boundary objects of academic scholarship and self-publishing; the legal fictions of fair use, the idea/expression dichotomy, and work for hire, as well as contractual arrangement affecting these areas.

The Impact of MIT's OpenCourseWare Model

Returning to the MIT OpenCourseWare model, it is remarkable that the actions outlined above as potentially most beneficial from a tactical viewpoint are precisely the opposite of the actual decisions made by the faculty and administration at MIT. Rather than opting for an individualistic model of information production and dissemination, the University made a conscious choice to support online scholarship and web information sharing, aligning their efforts squarely with a communal notion of the university as a resource for information sharing, not only within the academic community, but worldwide  [2]. The MIT decision can be viewed as a testament to the strength of institutional norms, whereby scholarly norms of information sharing overpowered the push of market-driven attempts to align the institution more closely with individualistic models of success. It will be interesting to see if, as predicted by complexity theory, the norms of information sharing that drive the OpenCourseWare project are followed up with punishments for those individuals who refuse to cooperate with the project. In the absence of punishments such as denial of tenure to non-cooperating staff members, theory predicts that defections would eventually undermine the project.

It is also interesting to note the decision to develop the OpenCourseWare model was framed by its supporters as consistent with, rather than contradictory to, the market-based initiatives of the university, even while the decision adhered to institutional norms. Arguing that the OpenCourseWare helped encourage alumni to provide further support for university activities, marketed the University to funders and prospective students as an innovative leader, and increased the exposure of future contractors to the University's offerings, indicates that the University was well aware of the potential objections to the project, and sought to minimize conflict by framing OpenCourseWare as serving broader market aims, while eschewing short-term market advantage.

Viewed more broadly, the decisions made at MIT have broad implications within the overall system of university research, service, and teaching, since it provides a counterintuitive model to the prevailing view of the university as an individual actor in a competitive marketplace. While not unilaterally rejecting the notion of the university as a market-situated institution, it reminds us that the value of the university to society lies equally in its ability to honor communal obligations for knowledge creation and dissemination.

What should be made clear is that the MIT model does not answer the boundary problems highlighted above. However, situating the MIT experience in our framework of information policy tectonics helps provide a checklist of information policies that need to be adjusted to accommodate the model. This is not a trivial task, since it presumes a mode of information exchange based on a gift economy rather than a market economy, a notion deeply inimical to current intellectual property law. Whether other institutions will seek or be able to follow MIT's lead in this endeavor will depend partly on the internal culture of the University and the resources available to manage the technical, legal, and social challenges of the model. More critically, however, for the MIT model to work from a systemic perspective, the establishment of more community-friendly global information policies will be required.

If we support the efforts of MIT and seek to replicate the model elsewhere, some outstanding information policy issues need to be resolved. Following are some suggestions for changes both to legislation and practice, based on an analysis of the trends outlined in Figures 2 and 3.

Deciding Whether Information is Property?

Overall, we need to decide whether information developed through the scholarship process should be considered property or not. The rhetoric about the property status of information has traditionally been equivocal. On one hand, we recognize that information content, unlike property, can be enjoyed by many people without the loss of value by any single individual. In fact, some scholars argue that information possesses strong network effects, where the value of the information increases when more people have access to it  [13]. We further recognize that new combinations of information are built primarily on previous works. Alternatively, most formats for information storage and presentation are both tangible and excludible, making them feasible for treatment as physical property. In fact, the legal system for information ownership is most commonly known by the boundary term 'Intellectual Property,' which links the creative work of authorship to the market language of property and ownership most common in private sector information production and dissemination.

If we accept the notion that scholarly work should be viewed primarily as property, then we might look to mechanisms for protection of physical public property, particularly property that is critical to our ecological health  [14]. In this scenario, mechanisms such as use of Eminent Domain (e.g. to ensure open access to the genome, or to information about AIDS fighting drugs), information conservancies and cooperatives, continuation of government grant programs, use taxes and credits, and impact statement requirements for individuals claiming ownership for information might be marshaled in defense of the commons.

If we believe that information is not primarily like property, then we must develop more subtle mechanisms for commons protection. Initiatives like MIT's rely on statements of faith by professors and decisions by universities to honor communal rather than individual norms of behavior. Absent legal and regulatory mechanisms to support and provide incentives for communal behavior, the effort is prone to free-rider problems and self-interested decision making. The enforcement of institutional norms in support of communal information sharing could be effective within the boundaries of universities, where some leverage exists, but only if universities align institutional norms to consistently support communal rather than individual behavior.

Alternatively, open source and open copyright licenses are useful but awkward contractual methods for establishing legal mechanisms for communal knowledge sharing. They require a large administrative overhead to enforce, and can be susceptible to attack if their results are perceived to wander too closely to activities over which private proprietors have claimed ownership [15].

Sharing Information on the Web

Web-based information sharing requires a policy model based on communal rather than individualistic norms. This means protection both against claims of individual ownership by other producers and disseminators, and mechanisms for active partnerships among disciplinary communities based communal knowledge sharing rather than individual authorship and inventorship. In the area of information use by universities for online education, legislative enactments like the United States' TEACH Act of 2002 (H.R. 2215), which are based on presumptions that a university will control access to course materials through registration and access controls, are ultimately not useful. They situate universities only within the market economy, and separate knowledge communities based on institutional boundaries. Utilizing institutions such as information cooperatives or research organizations mentioned above to support subscription-free scholarly publishing and archiving, rather than relying on private publishing models, might be a more appropriate commons-based response. These actions would undercut private publishers' moves to monopolize scholarly publishing and provide fee-based online courses.

Redefining Copyright's Public Domain

The return of an information commons in which the default condition for information texts is the public domain rather than copyright would also simplify information sharing. It was not until 1976 (with the removal of registration requirements) that the United States moved from a public domain to a copyright default position regarding texts. Patents have always required affirmative actions in order for protection to exist. Establishing a new default position for public domain copyrights would not necessarily mean a return to registration, which would be difficult to implement in the current global information policy environment. Rather, legislatures could codify a new type of copyright, in which an author presents a work to the public domain, while retaining rights to ensure that authorship is properly acknowledged and protected from distortion. This would provide a legal foundation for the norms against plagiarism and for citation.

Reinforcing Institutional Norms

At the University level, faculty evaluation and tenure processes that actively reward non-market collaboration, inter-disciplinarity, and commitment to communal projects both inside and outside the walls of the University will help establish incentives and norms for sharing. It should be realized that this is in direct conflict with current pressures many faculty and institutions feel to establish individual control over information assets, whether through copyrights, patents or contractual agreements.

Clarifying Ownership of Government-funded Information

On another front, information ownership rules regarding discoveries and information created through government granting programs have provided a mechanism to transfer inventions and discoveries made under government grant funding to universities and faculty, in that order of precedence (see the U. S. Bayh-Dole Act of 1980 (35 U.S.C. 200-212)). However, the process of patenting requires the modification both of the process of scholarship and the outcome of scholarly research from a communal focus based on pure research to an individualistic focus based on personal inventorship and product monopoly  [5]. Providing a legal mechanism to protect communal research from individualistic raids without the necessity of defining individual ownership or patent utility would help avoid the necessity of individualizing invention and discovery. Self-regenerating information cooperatives or research organizations could serve as stand-ins for individuals for legal purposes, if we were to export the legal 'work-for-hire' fiction from copyright to patent law.

Additionally, codifying current governmental grant rules regarding open data sharing in projects funded by government monies would help stem the push toward sui generis database protection. Current practice  [16] specifies that grantees have an affirmative obligation to make data collected with public funds available for general use after a limited period of time, during which grantees are expected to publicize their findings.

In summary, re-energizing the University's role as the primary locus for information and knowledge development and sharing from a society-wide perspective will require more efforts like MIT's OpenCourseWare. However, institutionalizing efforts such as MIT's will also require active debate and leadership to carve out a coherent legal and cultural regime distinct from the current market-based, property-driven system. The information policy balance originally developed in such documents as the US Constitution can no longer rely on the pretense that the public domain (envisioned as everything outside a limited system of property-like control over information) is large and robust enough to withstand persistent efforts to 'fence in' the commons.

The role universities can play in protecting the commons is twofold. First, working with governments and foundations, they can develop and energize new systems for information creation and sharing that serve as alternative models to the ascendant market systems. Secondly, they can partner with organizations like libraries, Open Source developers, and communities committed to open exchange of information both within and outside the United States to press for new legal and other protections for the commons.

REFERENCES

1. Walsh, C. (2003). MIT 32 Courses Online for Free: University Website Drew 42M Hits in First Month. Patriot Ledger, p. Education, 2.

2. Vest, C. M. (2004). Why MIT Decided to Give Away All its Course Materials via the Internet. The Chronicle of Higher Education, p. 20.

3. Coggins, P. (2001). MIT Initiative Raises Issue of IP Rights in Academia. The National Law Journal, p. C4-5.

4. Olsen, F. (2002). MIT's Open Window. The Chronicle of Higher Education, p. IT, p. 31.

5. McSherry, C. (2001). Who Owns Academic Work?: Battling for Control of Intellectual Property. Cambridge, MA: Harvard University Press.

6. Maxwell, T. A. (2003). Toward a Model of Information Policy Analysis: Speech as an Illustrative Example. First Monday, 8(6). http://www.firstmonday.org/issues/issue8_6/maxwell/index.html. Accessed 4/12/2004.

7. Star, S. L. and. J. R. Greismer. (1989). Institutional Ecology, 'Translations' and Boundary Objects. Social Studies of Science, 19(3), 387-420.

8. Tebbel, J. (1987). Between Covers: The Rise and Transformation of American Book Publishing New York: Oxford University Press.

9. Patterson, L. R. and S. W. Lindberg. (1991). The Nature of Copyright: A Law of User's Rights. Athens: University of Georgia Press.

10. Britannica (2004). Encyclopaedia Britannica 2004. [DVD].

11. Glendon, M. A. (1993). Rights Talk: The Impoverishment of Political Discourse. New York: Free Press.

12. Axelrod, R. (1997). The Complexity of Cooperation (anonymous, Ed.). Princeton: Princeton University Press.

13. Liebowitz, S. a. S. E. M. (2004). Network Externalities., 2004 Accessed 4/12/2004. http://www.utdallas.edu/%7Eliebowit/palgrave/network.html

14. Boyle, J. (2003). The Public Domain: The Second Enclosure Movement and the Construction of the Public Domain. Law and Contemporary Problems, 66(1), 33-74.

15. Vaughan-Nichols, S. (2004). IBM Goes After SCO's Copyright Claims. eWeek. http://www.eweek.com/article2/0,1759,1559009,00.asp accessed 4/9/2004

16. Reichman, J. H. and. P. Uhlir. (2003). A Contractually Reconstructed Research Commons for Scientific Data In a Highly Protectionist Intellectual Property Environment. Law and Contemporary Problems, 66 (Winter/Spring), 315-462.