College and Research Libraries The Strength of Weak Ties in Electronic Development of the Scholarly Com.munication System. Charles A. Schwartz How would knowledgeable use of the Internet develop within academic insti- tutions or various research fields and then diffuse across the loosely coupled scholarly communication system? Conversely, how might the scholarly system become balkanized into autonomous, even antagonistic, cultures or camps based on differing technological competencies and interests? Three overlapping mod- els of innovation (new technology) diffusion are described in relation to the Internet: individual threshold, critical mass, and the strength of weak ties. Two contrasting scenarios of a balkanized system are drawn: separate tables and braking mechanisms. The analysis discusses the prospective role of academic librarians in electronic development of the scholarly system and concludes with a note on future research in this area. [il ittle is known, in any sys- . tematic way, about the im- pact of the Internet on pat- terns of scholarly communi- cation. While the effects of computer networking in the workplace have re- ceived some attention over the past dec- ade, empirical research on the role of the Internet in the scholarly communication system hardly has begun. The literature in this area generally is marred not only by a lack of analysis but also by certain problems which have deterred under- standing. These include an ambiguity of Internet effects and incommensurable differences between the print and elec- tronic forms of communication. An ambiguity of Internet effects-func- tions (advantages) being inseparable from dysfunctions (disadvantages)-is readily apparent to anyone who has joined a few electronic groups: .To inhibit the flow of "useless junk" is to risk the loss of one of the most valuable impacts of computer-medi- ated communication systems-the flow of potentially useful information and ideas among persons with no pre- vious or off-line communication links.1 Bound up with that dilemma is a law of diminishing returns: the more infor- mation accessed, the less its overall meaning. In the midst of such ambiguity, two sharply divided schools of thought have emerged. Enthusiasts extol the revolutionary potential of the Internet to transmit interactive communication around the world at a rate close to the speed of thought.2 More conservative analysts, however, emphasize the limited experience of most scholars in using the Internet. In that behavioral perspec- tive, the "key issue in technological Charles A. Schwartz is Social Sciences Bibliographer at the Fondren Library, Rice University, Houston, Texas 77251-1892; e-mail: schwart@rice.edu. The author wishes to thank Gary Byrd, Martin Dillon, Stevan Harnad, David Knser, Sara Kiesler, Diane Kovacs, Charles Osburn, Prentiss Riddle, John Rutledge, and Herbert White for their helpful comments on an earlier draft of this paper. 529 530 College & Research Libraries innovation lies not within technol- ogy itself but among its potential us- ers-whether they possess a clear vision of opportunities that permit wise choices about what to pursue and what to prevent."3 On a more fundamental plane, the two schools have come to reflect incommen- surable paradigms whose bases include ev~rything from computer experience and scholarly tradition to time, space, and reality. Communication across that great divide is inevitably partial, for sus- tained and knowledgeable use of the In- ternet is a matter of personal conversion rather than simple logic: The computer's allure is more than utilitarian or aesthetic; it is erotic. In- stead of a refreshing play with surfaces, as with toys or amusements, our affair with information machines announces a symbiotic relationship and ultimately a mental marriage to technology. 4 Consider, for example, the two schools' split assumptions about human adjustment to technological innovation. Conservatives contend that "people are not easily sold on anything [like the In- ternet] that promises change in cogni- tive processes and organizational social structure."5 Enthusiasts take a different tack, that "as we've learned from the history of the telephone, radio, and tele- vision, people can adopt new communi- cation media and redesign their way of life with surprising rapidity."6 Or, consider mental models of the place of virtual reaHty. Where pioneers behold a "post-Gutenberg galaxy," other scholars discern an "infinite cage" in which the computer's faceless language and protocols threaten to govern the very processes of thought,7 .s Overall, it is not surprising that several writers have come to the view that few other areas of modern social science have such a large number of unsubstantiated speculations and such a small number of serious studies.9 A new field can be advanced if re- searchers agree on a framework of sig- nificant and feasible issues. Toward that end, this article suggests two allied is- sues: (1) the impact of the Internet on the structure of the scholarly communica- November 1994 tion system, and (2) the prospective role of academic librarians in that process. The scholarly system has a loosely coupled structure, one with remarkably little interaction, coordination, or even direct cause-and-effect relationships among the main constituencies: univer- sities, academic libraries, computing centers, publishing houses, the scholars themselves, and their societies. An es- sential issue is whether the Internet tends to make the system generally more tightly coupled (interdependent) or loosely coupled (organized anarchy). The allied issue is which group(s) will take a leading role in the dual process underlying new technology diffusion in the scholarly system: on a micro level, to mentor adoption and knowledgeable use of the technology; and on a macro level, to make the connections across groups-the strength of weak ties-on which system- wide diffusion of innovation depends. An essential issue is whether the Internet tends to make the system generally more tightly coupled (interdependent) or loosely coupled (organized anarchy). The thesis is that academic librarians are strategically situated to be the main agents of electronic development of the scholarly communication system. In that role, they can help prevent academic in- stitutions and other parts of the system from becoming unnecessarily divided into separate cultures or camps based on differing technological competencies and interests. The analysis focuses on informal pat- terns of online scholarly communica- tion. It does not cover the tangle of rudimentary issues involving develop- ment of peer-reviewed electronic jour- nals. That latter, more futuristic, topic is treated in the author's earlier work on scholarly communication as a loosely coupled system. 10 The article is arranged in four parts. The first describes scenarios of a balkan- ized system. The second part outlines a set of models of innovation (new tech- nology) diffusion that would lead to a more tightly coupled system. The next part focuses on the prospective role of academic librarians as the strength of weak ties in the loosely coupled system. The last part is a note on future research in this area. SCENARIOS OF A BALK.ANIZED SYSTEM If history is any guide, there is a real prospect of academic institutions or re- search fields breaking into autonomous, even antagonistic, cultures or camps based on differing technological compe- tencies and interests. The scenarios de- scribed below are not bound to become dominant but doubtless will be evident. Separate Tables In the 1960s computer statistical pack- ages associated with the behavioral movement caused tremendous divisive- ness in academic departments and scholarly societies. The result, described decades later by Gabriel Almond, is that "in some sense the various schools and sects now sit at separate tables, each with its own conception of proper sci- ence, but each protecting some secret island of vulnerability. 1111 Almond took his metaphor from Separate Tables, a 1955 play in which solitary diners in a hotel convey the loneliness of the human condition. Tables are a popular meta- phor; scholars in the humanities com- plain that they do not have a "place at the table," that they are "starving at the banquet. 1112•13 In this scenario, some functions of In- ternet groups-their international scope, nearly instantaneous interactive dynamic, and social equality-could generate a resurgence of chauvinistic conflicts over competing research agen- das, theories, and methods. Even Howard Rheingold, a dean of Internet pioneers, recognizes this prospect: The willingness of the online popu- lation to tolerate wide diversity of opinion might be ... an artifact of the early stages of the medium's growth. Fragmentation, hierarchization, rigidi- The Strength of Weak Ties 531 fying social boundaries, and single- niche colonies of people who share intolerances could become prevalent in the future.14 Separatism could take the broader form of a counter ethos to the Internet as a symbol of intelligence and modernity if numbers of scholars become appalled by computer jargon, electronic junk, and the semblance of "systems people" to a new-age "priesthood." A counter ethos might also develop in reaction to a fail- ure of some hypertext systems to meet scholarly needs.15 Braking Mechanisms An analogy drawn from the transfor- mation of the former Soviet system high- lights the sheer difficulty of cultural and behavioral change in any large-scale so- cial system. Stevan Hamad coined the term intellectual perestroika to signify a restructuring of the pursuit of knowl- edge in the electronic era.16 Other writers describe such fundamental change in a similar fashion: Adoption [of the Internet] has en- tailed a rather difficult process of unscrambling old procedures and at- titudes, moving to new ways of per- forming intellectual tasks and of thinking about communications, and then installing the new processes into the daily agenda of individuals and groups.17 Back in the U.S.S.R., when the Soviet form of perestroika began to deteriorate in the late 1980s, Mikhail Gorbachev complained that traditional institutions and ways of thinking were operating as "braking mechanisms" on the reform movement.18 In the West, certain properties of either the Internet or the traditional scholarly system thus far have operated as braking mechanisms on electronic progress. These include perceptions of cognitive over- load and information overload, lack of identifiable productivity gains, and lack of academic rewards for scholars to use the Internet. Cognitive Overload. The Internet is often likened to the ancient Library of Alexandria, which had a world of I 532 College & Research Libraries information but little in the way of a card catalog. As conservatives point out, navigating the Internet involves funda- mentally new skills. Though not "rocket science," such skills do require a com- mitment to learn and pose continuing frustrations over inconsistent protocols, redundant or incomplete search-engine retrievals, incompatible text formats, and so forth. While such frustrations continually are being alleviated by tech- nical refinements, the problem of cogni- tive overload should be seen in a broad context. The architectural scheme of the Internet remains that of ARPANET, the first computer network designed in the 1960s as a "doomsday" device-a com- munications and command medium that could survive a nuclear war by vir- tue of having no central control on either a policy or a technical level. Enthusiasts, for their part, are not re- ally concerned about technical difficul- ties: "We are early adopters of a chaotic technology, and the momentum of our own enthusiasm generally carries us be- yond the many annoyances and impedi- ents that stand in the way of an easy, natural, information retrieval environ- ment."19 Indeed, some librarians make a game out of difficult search and re- trieval in the form of Internet "treasure hunts," a behavior that anthropologists would term "galumphing"-the volun- tary placing of obstacles or complica- tions in one's path because the center of interest is process rather than goal.20 Information Overload. The Internet expedites a host of traditional scholarly needs: current awareness of professional developments, exchange of information on a timely basis, and collaboration among distant colleagues. A new func- tion is the creation of online journals that are "laboratories rather than show- cases," enabling a shift in scholarly com- munication from a finished product to the process of developing "knowledge in conversation."21 Enthusiasts make the claim that, "once we're all connected," the Internet will be a liberating, edifying experience of global proportions. Some collection development policies for the Internet even call for libraries to main- November 1994 tain archives of electronic groups (not just journals) as part of the "scholarly record."22 Conservative analysts, however, have a darker vision of the Internet being ·filled with "unmitigated garbage: off- the-cuff ideas, rabid diatribes, ideologi- cal vendettas-topics and recreations that have little relevance to any commer- cial, scientific, or serious activity. "23 Sci- ence fiction writers of the cyberpunk genre address this problem in different moods. Some writers depict a new pro- fession of knowledge mediators called hackers. In the next century, there are so many computer sites, networks, and da- tabases that accessing the Internet, in one novelist's delightful hyperbole, has become equivalent (in paper format) to "arranging for a 7 47 cargo freighter filled with telephone books and encyclo- pedias to power-dive into one's unit every couple of minutes, forever." 24 Other writers discount the role of knowledge mediators (whether hackers or librarians) to manage information overload. In a novel set in the year 2038, the Internet has become a "rowdy babel, a torrent of confusion and comment, made worse because in order to be no- ticed each user sends out countless cop- ies of his messages to any node that might conceivably listen." A courtesy monitor warns people not to "act like mental patients who shout out anything that comes to mind."25 The contemporary impact of the In- ternet on scholarly communication probably varies with each of the 1,200 or so discussion groups, as well as with individual competencies and interests. Still, the thrust of the Internet is hardly scholarly. As Charles McClure and asso- ciates found in a series of surveys, schol- ars have a clear aversion to online journals because "electronic publication does not enhance one's status or image; in fact, it may very well harm them." 26 The few successful attempts at estab- lishing scholarly (indexed) online jour- nals have relied on a strategy of putting famous researchers on editorial boards and having them twist the arms of col- leagues to submit papers.27 The thrust of the Internet is the socia- bility of networking. Indeed, a unique feature of this fiercely egalitarian me- dium is its diminished social-status ef- fects: "People who regard themselves as physically unattractive report feeling more lively and confident when they ex- press themselves over the network. Oth- ers who have soft voices or small stature report that they no longer have to strug- gle to be taken seriously."28 The New Yorker captured this feature in a cartoon of a dog sitting at a computer terminal explaining to a puppy, "On the Internet, nobody knows you're a dog."29 We simply do not know whether the Internet will lead (at least in our lifetimes) to a grand, across-the-board renegotiation of historical print-age patterns of influence and interaction. By and large, people grapple with in~ formation overload (or simplify choice situations) by reducing environmental scanning; narrowing attention spans; and devising other rule-of-thumb strate- gies. For e-mail in particular, informa- tion overload tends to have a curvilinear pattern. Individuals with an intermedi- ate range of experience-about 20-50 online hours-are most susceptible to overload.30 With greater experience, in- dividuals tend to withdraw from some groups and develop better networking skills. The essential point is that, as func- tions slide into dysfunctions, optimal computer networking requires some ex- posure to information overload. Productivity Gains versus Opportu- nity Costs. To consider the prospective impact of the Internet, a two-level per- spective is helpful. Technology can have first-level efficiency effects and second-level social effects. First-level effects involve the use of new technol- ogy to do old things in better or faster ways. Second-level effects lead people to do new things, to pay attention to differ- ent things, to interact with one another differently, or to develop new needs or expectations. Second-level effects are ex- The Strength of Weak Ties 533 traordinarily difficult to predict and emerge in society rather slowly, as peo- ple renegotiate outworn patterns of in- fluence and interaction.31 The literature on the Internet contains a few reports of first-level effects. For example, an online version of the Men- delian Inheritance in Man database, to which geneticists can post electronically a research note to a particular entry, is one way that networking can make a previously established form of scholarly communication more efficient.32 What is absent from the literature, however, is evidence of second-level effects of net- working on scholarly communication as a social system. We simply do not know whether the Internet will lead (at least in our lifetimes) to a grand, across-the-board renegotiation of historical print-age pat- terns of influence and interaction. Productivity gains from Internet ac- tivity are especially hard to specify. Even people with a lot of online experience tend to have only an intuitive grasp of its effect on their work. As the McClure team found in surveys, "members of groups have difficulty articulating spe- cific network impacts [even though] they cannot imagine working without networks. "33 Common responses to other surveys on the Internet's social ef- fects are likewise abstract: "increasing the stock of ideas" and "exchanging opinions. "34 Enthusiasts rightly contend that argu- ments about our inability to specify sec- ond-level effects of the Internet are too abstract to be really persuasive. Net- working's power to transform and mul- tiply the relativity of human interaction is plain enough. Still, the computer-pro- ductivity paradox and the distinction between efficiency and social effects re- main interesting areas for research. MODELS OF INNOVATION DIFFUSION LEADING TO TIGHTER COUPLING Up to now, this article has consid- ered prospects for electronic develop- ment of the scholarly communication system at the level of an individual scholar or librarian who must balance, 534 College & Research Libraries by experience or intuition, the benefits of learning how to navigate the Internet against the opportunity costs to other, more established professional concerns. As the McClure team found in surveys, scholars seeking access to the Internet typically have only one piece of research in mind and do not want to take a com- puter course or even deal with "systems people" to learn how to get the informa- tion. Instead, they need a personal con- sultant who understands their project well enough to offer application-specific training and "one-on-one hand-hold- ing."35 Other observers have drawn a similar conclusion-that academic li- brarians may find a unique niche as knowledge mediators who combine technical and disciplinary skills to meet specialized research needs. 36 Another two-level perspective is help- ful. The foregoing analysis is a mi- crolevel, somewhat static view of the scholar's teachable moment, when a particular need for the Internet arises. On a macrolevel, what are the conditions under which scholars' knowledgeable use of the Internet would be dynamic, spreading within academic institutions or across various research groups, then becoming a self-sustaining process in the system as a whole? In other words, how do individual decisions to adopt the Internet (or any innovation) possibly interact and aggregate in a loosely cou- pled system? Three overlapping models of collec- tive behavior address these questions at a systems level: individual threshold, critical mass, and the strength of weak ties. Although these models are not pre- dictive (they can be "proven" only after they occur), they have attracted wide attention as conceptual schemes that in- tegrate microlevel individual prefer- ences with macrolevel patterns of innovation diffusion. Threshold Model This model applies to a situation in which an individual has two alterna- tives: to do or not do a certain thing. The "thing" can be any binary decision in which one's choice depends, in part, on November 1994 the choices of some relevant group of individuals in the preceding period. Such decisions involve adopting-or re- jecting-an innovation, engaging in a political activity (striking, voting, or ri- oting), migrating, or conforming in vari- ous ways. The concept of threshold refers to the number or proportion of others who must make a choice before a given individual comes under the influ- ence of "bandwagon effects."37 As an illustration, individual prefer- ences regarding the Internet can be imagined on a 100-point scale. Pioneers have low thresholds (barriers to innova- tion); indeed, people like Charles Bailey Jr., Elaine Brennan, Diane Kovacs, or Stevan Harnad who start electronic jour- nals when almost everyone else is doing nothing of the sort have a threshold around zero. Edward Jennings, reflect- ing on the founding of Efournal, said that "given the efficiencies of the medium we were celebrating so noisily, it seemed to have taken us a ridiculously long time just to find a few people willing to listen seriously to our ideas."38 At the other end of the scale, conser- vatives-whether from fear, indiffer- ence, or lack of a mentor-have high thresholds. Actually, this model does not apply where most thresholds are very low or very high-that is, where an indi- vidual's behavior is not contingent on that of others. Thresholds in the middle range are the relevant ones in making the conceptual link between individual preferences and aggregate outcomes. Critical Mass This pertains to the turning point when an innovation is adopted-or re- jected-by enough individuals to in- duce many others to do the same.39 The critical ("take off") stage of Internet adoption started in 1987 when ex- tremely rapid advances in supercon- ducting research pointed up the inadequacy of preprints and telephones for scholarly communication; clusters of researchers adopting technology more or less simultaneously are a familiar phenomenon.40 However, the literature on computer networking draws a bleakly different scenario in which sus- tained use of the Internet spreads for a while, then recedes, leaving academic institutions and research fields di- vided into separate camps.41 In face of that prospect, discussions on upgrading the Internet to a national information infrastructure have included the need for equity policies which would serve to universalize access to networking infor- mation and services.42 The Strength of Weak Ties This model describes how an innova- tion can spread from group to group in a loosely coupled system.43 As opposed to bandwagon effects at the group level, the strength of weak ties focuses on the role that outsiders play in the diffusion process at the systems level. An indirect but striking example-pertaining to in- formation diffusion rather than techni- cal knowledge diffusion-comes from experiments on the French national net- work that uncovered the role of "cross- pollinators of telematique": As the [French] system evolved, it became a very loosely coupled collec- tion of different information services and communication forums. Many people stayed in only one or two dif- ferent domains, but a small number of people seemed to move ideas very swifly from one group to another. We found that we could feed a small piece of deliberately false information to one of these people, and it spread throughout all the different groups, to as many as four thousand people within two days.44 In a nutshell, outsiders are those indi- viduals who are most likely to deal with acquaintances in other groups (special- ties, organizations, or other branches of the same organization). Scholarly innovations tend to come from interdisciplinary patterns of inter- action at the margins of established fields. 45 If the innovativeness of a re- search (or social) group is shackled by vested intellectual (or cultural) interests, then new ideas must emanate from out- siders in the network of individuals. Weak ties thus provide the basis for both The Strength of Weak Ties 535 microlevel change (by broadening group horizons) and macrolevel integration (by expanding intergroup connections). What types of individuals are poten- tial agents of Internet use in the schol- arly communication system? Generally, it would be those who have ties that overlap generational, occupational, or knowledge groups: "Gaps in network at- titudes and skills exist along several di- mensions: between older and younger researchers, between researchers and network administrators, between peo- ple in different sectors, between re- searchers from different disciplines, and between researchers working on different kinds or different stages of problems."46 Success for a scholar as an Internet pioneer can be problematic. On the one hand, "some junior researchers fear that their network expertise would relegate them to a 'computer ghetto,' in which they no longer participate in the conceptual aspects of research. " 47 On the other, as the protagonist in a cyberpunk novel reflects when ex- ploring an outdoor music and technology fair: "Interesting things happen along borders-transitions- not in the middle where everything is the same. There may be some- thing hap- pening along the border of the crowd, back where the lights fade into the shade of the overpass."48 PROSPECTIVE ROLE OF ACADEMIC LIBRARIANS In considering the prospect of the scholarly communication system becom- ing balkanized into separate cultures or camps, one must weigh the opportunity costs of learning to use the Internet against other professional concerns. A vast ma- jority of scholars may simply go along with what tenure committees recognize- teaching and publishing as usual-and avoid what they perceive to be a techno- logical hassle. Certainly, there is no shortage of warnings by conservatives against "mindless safaris into galaxies of informational garbage."49 The idea that the library should as- sume a leading role on campus in devel- oping positive faculty attitudes about 536 College & Research Libraries the use of new technology is hardly a new one. My point concerns a broader, more systematic, even historic process. Academic librarians, by facilitating knowledgeable Internet activity-teach- ing short courses; publishing descrip- tive accounts in scholarly journals; and making alliances, under the auspicies of the As~ociation of College and Research Libraries, with scholarly societies-can provide the strength of weak ties on which systemwide adoption and inte- gration of the new technology rests. Field Variances in the "Strength of Weak lies" Model The importance of this prospective role will vary, as will patterns oflnternet adoption, with the nature or structure of collegial interaction within a particular field. Such interaction differs mark- edly among fields. In science and tech- nology fields, work tends to be highly . collaborative within a department be- cause colleagues have a common envi- ronment-they share the same technology and much professional knowledge. Thus, one would expect departmental ties to be the primary social influence for scien- tists to adopt the Internet. Such close influence can be called the "classical Athens interface. "50 The social sciences and humanities, by contrast, have much less opportu- nity for collaboration within a depart- ment, partly because faculty hiring is geared to maximizing intellectual diver- sity as a means of ensuring broad instruc- tional coverage. In that kind of setting, given the lack of shared technology and expertise on campus, collegial support tends to be on a regional or national "invisible college" leveJ.51 Survey Research on Field Variances Internet surveys conducted by the writer in 1993, while not having a rigor- ous level of statistical reliability (in the 90 percent range of confidence that responses are not merely random), are nonetheless broadly supportiv,e of these alternate theories. For respon- dents of CIVIL-L (Civil Engineering Re- search and Education), one-third were November 1994 prompted to adopt the Internet by de- partmental colleagues and another third by more distant contacts (librarians, computer specialists, or "invisible col- lege" acquaintances); the last third were self-starters. In contrast, 80 percent of respondents of HUMANIST (Humani- ties Computing) were prompted by dis- tant social influences to adopt the Internet and another 10 percent by de- partmental influences; the remaining 10 percent were self-starters. The problem of high statistical reli- ability is that Internet surveys tend to have very low response rates, 5 percent or less. 52 The two surveys described here had rates of about 20 percent (n=30), but that rate was accomplished by asking only one question and by sending it per- sonally addressed to individuals instead of posting it, bulletin-board style, on a listserver. Low response rates are indicative of task-centered groups, which are charac- terized by strong individualism, low so- cial cohesion, mobility of membership, and relatively narrow goals. Such groups, especially online, are well suited to enhance the information-gathering functions of relationship-centered groups, which have broad mandates, stable memberships, and holistic personal rela- tionships.53 Yet, the relatively narrow purview of online scholarly groups- with their low response rates to Internet surveys-is an important qualifier to the idea of "virtual communities." FUTURE RESEARCH At a broad, systems-level analysis, the Internet will tend to balkanize the schol- arly communication system into sepa- rate camps or cultures based on differing technological competencies and inter- ests. Of great interest, however, will be all the exceptions to that generalization. The force of some braking mechanisms- cognitive overload and information over- load-will depend in part on the structure of collegial interaction within a particular field. The force of other braking mecha- nisms-lack of clear productivity gains or academic rewards-might lessen, over time, if computer networking becomes a very symbol of intelligence and moder- nity. Simply put, the Internet will affect the various fields in diverse ways and at different rates. Comparative case studies will reveal a central paradox of our time, that the scholarly communication system is be- coming simultaneously more unified (tightly coupled) and more fragmented (loosely coupled). A rudimentary hy- pothesis is that the growth of electronic journals and groups in centralized fields will have decentralizing effects, whereas such growth in decentralized fields will have centralizing effects. 54 Another area for comparative case study involves patterns of social influ- ence in new technology adoption and diffusion. The strength of weak ties model proposed here is different from the conventional model of "integrated" The Strength of Weak Ties 537 librarianship. In the latter model, which evolved in the 1970s, subject specialists staff the reference desk and serve all comers, including those with simple in- formation needs. While that model had fallen into "conceptual disarray" aerry Campbell's phrase) by the late 1980s, the Internet should accelerate its demise. Just as scholars face opportunity costs (e.g., to research productivity) in learn- ing how to navigate the Internet, aca- demic librarians must be relieved of labor-intensive tasks if they are to have a more sophisticated involvement with the new technology, particularly that on UNIX-based systems. This is an impor- tant shift, one that warrants analysis and understanding, for it has aroused a strong debate that reflects the emer- gence of antagonistic cultures or camps within our own profession. 55 REFERENCES AND NOTES 1. Starr Roxanne Hiltz and Muray Turoff, "Structuring Computer-Mediated Communication Systems to Avoid Information Overload," Communications of the ACM 28 Guly 1985): 681. 2. Stevan Hamad, "Post-Gutenberg Galaxy: The Fourth Revolution in the Means of Production of Knowledge," Public-Access Computer Systems Review 2, no. 1 (1991): 39-53; e-mail retrieval: Listserv@uhupvml.uh.edu; Get HARNAD PRV2N1 f=mail. 3. Carolyn C. Lougee, "The Professional Implications of Electronic Information," Technology, Scholarship, and the Humanities: The Implications of Electronic Information, Getty Art History Information Program, Sept. 30./0ct. 2, 1992, anonymous ftp: FTP.CNI.ORG; cd/ CNI/documents/tech.schol.human/papers; Binary; Get LOUGEE.WP. 4. Michael Heim, "The Erotic Ontology of Cyberspace," in Cyberspace: First Steps, ed. Michael Benedikt (Cambridge, Mass.: MIT Pr., 1991), 61. 5. Hiltz, "Structuring," 688. See also Langdon Winner, "How Technology Reweaves the Fabric of Society," Chronicle of Higher Education, Aug. 4, 1993, B1-B3. 6. Howard Rheingold, "A Slice of Life in My Virtual Community," in Global Networks: Computers and International Communications, ed. Linda M. Harasim (Cambridge, Mass.: MIT Pr., 1993), 56. 7. Hamad, "Post-Gutenberg." 8. Michael Heim, The Metaphysics of Virtual Reality (New York: Oxford Univ. Pr., 1993), 79. 9. Charles R. McClure and others, The National Research and Education Network (NREN): Research and Pflicy Perspectives (Norwood, N.J.: Ablex, 1991), chapter 7 [hereafter referred to as NREN] f Stephen Gould, High Performance Computing: An Overview (Washington, D.C.: Library of Congress, Congressional Research Service, 1990), chapter 1/ Charles Henry and Paul Evans Porter, "Networks and Scholarly Commurication," Bulletifi of the American Society for Information Science 19 (Feb./Mar. 1993):17 ~tephen R. Ruth and Raul Gouet, "Must Invisible Colleges Be Invisible? An Approach to ExftiD-ining Large Communities of Network Users," Internet Research 3 (Spring 1993): 36-53/ James G. March, "Old Colleges, New Technology," in Computing and Change on Campus, ed. Sara Kiesler and Lee Sproull (New York: Cambridge Univ. Pr., 1987), 16-27. 10. Charles A. Schwartz, "Scholarly Communication as a Loosely Coupled System: Reassessing Prospects for Structural Reform," College & Research Libraries ~5 (Mar .. 1994): 101-17. 538 College & Research Libraries November 1994 11. Gabriel A. Almond, A Discipline Divided: Schools and Sects in Political Science (Newbury Park, Calif.: Sage, 1990), 13. 12. Lawrence Dowler, "The Implications of Electronic Information for National Institutions," Technology, Scholarship, and the Humanities: The Implications of Electronic Information [see reference 3] Get OOWLER.WP. 13. Barrett S. Caldwell and Lilas H. Taha, "Starving at the Banquet: Social Isolation in Electronic Communication Media," Interpersonal Computing and Technology: An Electronic Journal for the 21st Century 1 Qan. 1993). Anonymous ftp: GUVM.CCF.GEORGETOWN.EDU; Logon IPCT-J; Password is GUEST; GET CALDWELL IPCTV1N1. 14. Howard Rheingold, The Virtual Community: Homesteading on the Electronic Frontier (Read- ing, Mass.: Addison-Wesley, 1993), 207. 15. One example involves Thesaurus Linguae Graecae, a computer disk containing nearly all classical Greek texts (some 62 million words). It has not had much of an impact on scholarship because many so-called "imported concepts" that are useful for understanding classical Greek society-e.g., sovereignty, state, ideology, citizenship-are not part in the ancient Greek language. W. R. Connor, "Scholarship and Technology in Classical Studies," in Scholarship and Technology in the Humanities, ed. May Katzen (New York: Bowker Saur, 1991), 58-60. In a similar vein, Oleg Grabar shows how poorly designed computer products can result when technicians who design them do not understand certain field-specific subleties. Oleg Grabar, "The Intellectual Implications of Electronic Information," Technol- ogy, Scholarship, and the Humanities: The Implications of Electronic Information [see refer- ence 3] Get GRABAR.WP. 16. Steven Hamad, "Scholarly Skywriting and the Prepublication Continuum of Scientific Inquiry," Psychological Science 1 (Nov.1990):342. 17. R. William Maule, "Infrastructure Issues in Computer-Mediated Communication," Elec- tronic Journal of Communication 3, no. 2 (1993): 2. E-mail retrieval: Com- serve@vm.its.rpi.edu; Send MAULE V3N293. 18. Mikhail Gorbachev, Perestroika: New Thinking for Our Country and the World (New York: Harper & Row, 1987), 19. 19. Stuart Wiebel, "Mime and the Future of Internet Journals," Archnet Electronic Journal on Virtual Culture 1, no. 5 (1993). E-Mail retrieval: Listserv@kentvm.kent.edu; GET SQARV1N5 Weibel. 20. Stephen Miller, "Ends, Means, and Galumphing: Some Leitmotifs of Play," American An- thropologist 75 (Feb. 1973): 87-98. 21. Eyal Amiran, "Refereed Electronic Journals and the Future of Scholarly Publishing," in Advances in Library Automation and Networking 4 (Greenwich, Conn.: JAI Pr., 1991), 35-36. 22. Committee on Institutional Cooperation, "Task Force on the CIC Electronic Collection," Oct. 1, 1993, unpublished manuscript. 23. Abridged from Tom Forester and Perry Morrison, Computer Ethics: Cautionary Tales and Ethical Dilemmas in Computing, 2d ed. (Cambridge, Mass.: MIT Pr., 1994), 255. 24. Neal Stephenson, Snow Crash (New York: Bantam, 1992), 20. 25. David Brin, Earth (New York: Bantam, 1990), 289, 518. 26. McClure, NREN, 103. 27. David L. Wilson," A Journal's Big Break," Chronicle of Higher Education, Jan. 26, 1994, A23. 28. Lee Sproull and Sara Kiesler, "Computers, Networks, and Work," Scientific American 266 (Sept. 1991):120. Counterpoint to this fierce egalitarianism is the need for the scholarly system to develop new norms of communication: "We are still a community that is learning how to talk to itself, a community that is learning how to describe itself, and a community that is trying to learn the standards that will allow us to talk to one another without friction or noise." Elaine Brennan, "Informal Publication and the Scholarly Record: Bits and Bytes from the Experience of Editing HUMANIST and Other Electronic Lists," in Scholarly Communication in an Electronic Environment, ed. Robert Sidney Martin (Chicago: ALA, 1993), 44. 29. P. Steiner, cartoon, New Yorker 69, July 5, 1993, 61. 30. Hiltz, "Structuring," 681. 31. Sara Kiesler and Lee Sproull, "Group Decision Making and Communication Technology," Organizational Behavior and Human Decision Processes 52 (1992): 96-112; Clifford Lynch, The Strength of Weak Ties 539 "The Transformation of Scholarly Communication and the Role of the Library in the Age of Networked Information/' Serials Librarian 23, no. 3-4 (1993): 5-20. 32. Richard E. Lucier, "Knowledge Management Refining Roles in Scientific Communication," EDUCOM Review 25 (Fal11990): 21-27. 33. McClure, NREN, 101. 34. Starr Roxanne Hiltz, Online Communities: A Case Study of the Office of the Future (Norwood, N.J.: Ablex, 1984), 180. 35. McClure, NREN, 156. 36. Erwin K. Welsch in Charles W. Bailey, Jr. and Dana Rooks, ed., "Symposium on the Role of Networked-Based Electronic Resources in Scholarly Communication and Research," The Public-Access Computer Systems Review 2, no. 2 (1991): 4-60. E-mail retrieval: Mail List- serv@uhupvm1.uh.edu; GET BAILEY PRV2N2 F=mail. Paula T. Kaufman and others, "Scholarly Communications: New Realities, Old Values," Library Hi Tech 10, no. 3 (1992): 61-78. Thomas G. Kirk, "Beyond Information Sources: The Spirit of Networking," guest editorial in College & Research Libraries 55 Gan. 1994): 7-8. 37. Mark Granovetter, "Threshold Models of Collective Behavior," American Journal of Sociol- ogy 83 (May 1978): 1420-43. 38. Edward M. Jennings, "EJournal: An Account of the First Two Years," Public-Access Computer Systems Review~, no. 1 (1991): 97. E-mail retrieval: Mail Listserv@uhupvm1.uh.edu; GET JENNINGS PRV2N1 F=mail. 39. M. Lynn Markus, "Toward a 'Critical Mass' Theory of Interactive Media: Universal Access, Interdependence, and Diffusion," Communication Research 14 (Oct. 1987): 491-511. 40. Melvin Kranzberg, "Interdependence of Scientific and Technological Information and Its Relation to Public Decision Making," Telescience: Scientific Communication in the Informa- tion Age,l_he Annals of the American Academy ofPolitical and Social Science 495 Oan. 1988):37. ~~~~e first article on the prospective role of the Internet in scholarly communication is now a classic: J.C.R. Licklider, Robert Taylor, and Evan Herbert, "The Computer as a Communication Device," International Science and Technology (Apr. 1968):21-31. Yet the Internet's potential went virtually unnoticed for two decades; as late as 1986 Internet researchers were "uncertain if, how and why average users in post-secondary institutions would want to use large distributed mail networks." Geoffrey Gurd and Jocelyne Picot," A Study of Atlantic Canadian User Reactions to Two Inter-University Electronic Networks," Canadian Journal of Information Science 11 (1986): 102-21. A 1985 novel about the "global campus" focuses on FAX machines and jet airplanes; computer telecommunication net- works are not mentioned. David Lodge, Small Worlds: An Academic Romance (New York: Penguin, 1985), 43-44. 41. Vartan Gregorian, "Technology, Scholarship, and the Humanities: The Implications of Electronic Information," Technology, Scholarship, and the Humanities: The Implications of Electronic Information [see reference 3] Get GREGORIAN.WP. Markus, "Critical Mass," 492. Hiltz, Online Communities, 9. 42. Brian Kahin, "Overview: Understanding the NREN," in Building Information Infrastructure- Issues in the Development of the }!_ational Research and Education Network, ed. Brian Kahin (New York: McGraw-Hill, 1992), 13f~ouglas Greenberg, "You Can't Always Get What You Want: Technology, Scholarship, and Democracy," in New Technologies and New Directions, ed. G.R. Boynton ~Sheila D. Creth (Westport, Conn.: Meckler, 1991), 11-25. McClure, NREN, 161. 43. MarkS. (;f~novetter, "The Strength of Weak Ties," American Journal of Sociology 78 (May 1973): 1361-80; see also Mark Granovetter, "The Strength of Weak Ties: A Network Theory Revisited," in Social Structure and Network Analysis, ed. Peter V. Marsden and Nan Lin (Newbury Park, Calif.: Sage, 1983), 105-30. 44. Rheingold, Virtual Community, 228. 45. Thomas S. Kuhn, The Structure of Scientific Revolutions, 2d ed., enlarged (Chicago: Univ. of Chicago Pr., 1970), 460f? aryl E. Chubin, "The Conceptualization of Scientific Specialties," Sociological Quarterly 17 (Autumn 1976):448-76;iNoah Friedkin, "A Test of the Structural Features of Granovetter's 'Strength of Weak Ties1 Theory," Social Networks 2 (1980): 411-22. 46. McClure, NREN, 156. 47. Ibid. 540 College & Research Libraries November 1994 48. Stephenson, Snow Crash, 113. 49. See, for example, A. Richard Turner, "Lights Are On, Will Anyone Be Home?" in Scholars and Research Libraries in the 21st Century (New York: American Council of Learned Societies, 1990), 27-30. For an opposing view, see Thomas W. Loughlin, "Virtual Relation- ships: The Solitary World of CMC," Interpersonal Computing and Technology: An Electronic Journal of the 21st Century 1 ijan. 1993); e-mail retrieval: Listserv@guvm.georgetown.edu; GET LOUGHLIN IPCTV1N1. 50. Derek Leebaert and Tunothy Dickinson, "A World to Understand: Technology and the Awakening of Human Possibility," in Technology 2001: The Future of Computing and Communications, ed. Derek Leebaert (Cambridge, Mass.: MIT Pr., 1991), 293-321. 51. Comparative studies of research fields give particular attention to the differential role of "insiders" and "outsiders": Stephan Fuchs, The Professional Quest for Truth: A Social Theory of Science and Knowledge (Albany, N.Y.: State Univ. of New York Pr., 1992); Yoram Neumann and Edith Finaly-Neumann, "The Support-Stress Paradigm and Faculty Research Publica- tion," Journal of Higher Education 61:565-80 (Sept./Oct. 1990), 120-43. 52. Personal communications with Diane Kovacs, Kent State Univ., on Sept. 13, 1993, regarding her survey work forthcoming in the Journal of the American Society of Information Science; and with Lewis 1)rler, Latin American Scholarship Program of American Universities, Harvard Univ., on Sept. 23, 1993. 53. Jan Walls, "Global Networking for Local Development: Task Focus and Relationship Focus in Cross-Cultural Communication," in Global Networks, 153-65. MarkJ. Schaefermeyer and Edward H. Sewell, Jr., "Communicating by Electronic Mail," American Behavioral Scientist 32 (Nov./Dec. 1988): 112-23. 54. George P. Huber, "A Theory of the Effects of Advanced Information Technologies on Organizational Design, Intelligence, and Decision Making," Academy of Management Re- view 15 (1990):57. 55. For analysis, see Jerry D. Campbell, "Shaking the Conceptual ~undations of Reference: A Perspective," Reference Services Review 20 (Winter 1992): 29-35, and Rebecca R. Martin, "The Paradox of Public Service: Where Do We Draw the Line?" College & Research Libraries 51 ijan. 1990):20-26 For debate, see Larry R. Oberg, "Rethinking Reference: Smashing Icons at Berkeley," College & Research Libraries News 54 (May 1993): 265-66;/and Daniel F. Ring, "Searching for Darlings: The Quest for Professional Status," College & Research Libraries News 54 (Dec. 1993):641-43.