Having Your Cake and Eating it Too: A Recipe for a Collaborative CWIS in a Decentralized Environment Copyright CAUSE 1994. This paper was presented at the 1994 CAUSE Annual Conference held in Orlando, FL, November 29- December 2, and is part of the conference proceedings published by CAUSE. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, that the CAUSE copyright notice and the title and authors of the publication and its date appear, and that notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To copy or disseminate otherwise, or to republish in any form, requires written permission from CAUSE. For further information: CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301; 303-449-4430; e-mail info@cause.colorado.edu HAVING YOUR CAKE AND EATING IT TOO: A RECIPE FOR A COLLABORATIVE CWIS IN A DECENTRALIZED ENVIRONMENT Lee Watkins, Jr. Assistant Director of Academic Computing Laura O'Callaghan Associate Director of News and Information The Johns Hopkins University Baltimore, MD ABSTRACT As a result of a unique partnership between four differing units at Johns Hopkins- Academic Computing, the Eisenhower Library, News & Information, and Student Affairs-a flexible, comprehensive, and inclusive campus-wide information system is now fully developed and highly successful at Johns Hopkins less than a year after inception. JHuniverse, as it known, has knit together faculty, students, staff and alumni across the divisions of the university and around the country in a manner that simply did not exist before, while providing the flexibility and "ownership" that is necessary to remain true to the spirit of entrepreneurial dispersion which characterizes Hopkins. By leveraging the strengths of each organization and capitalizing on the entrepreneurial nature of the institution, the project team was able to implement the system at very low cost while maintaining the commitment to provide universal free access in what is normally a pay- as-you-go internal economy. The result is one of the few true universitywide resources, rather than being perceived as an enterprise of the computing center, the library, or the administration. This presentation describes the genesis and development of-and prognosis for the interdivisional collaboration, but it is also the story of how JHuniverse has helped make decentralization a win-win proposition for Hopkins. BACKGROUND The Johns Hopkins University is a remarkably decentralized, geographically dispersed research university composed of eight very independent schools on three main campuses in two cities, a major research division in suburban Maryland, with numerous centers and affiliates in the Baltimore-Washington region and two in foreign countries. Given such an environment, a comprehensive networked information system would seem to be absolutely essential, yet differing priorities long conspired to prevent such a system from coming into existence. CAUSE attendees might find it difficult to believe that less than a year ago a worldrenowned institution like Johns Hopkins did not have an official campus-wide information system (CWIS). This is especially striking because these systems are now commonplace among universities and colleges. However, anyone familiar with Hopkins' historical emphasis on individual excellence and resistance to just about anything resembling centralized control would not be surprised in the least. This is part of a long tradition extending back to the first president of Johns Hopkins, Daniel Coit Gilman, one of the seminal minds in the history of American higher education. In his installation address, Gilman stated Hopkins' goals as "The encouragement of research ... and the advancement of individual scholars, who by their excellence will advance the sciences they pursue, and the society where they dwell." This simple vision is in fact what gave rise to the modern American research university as we know it today. It has served Hopkins exceptionally well, but it also resulted into an institution firmly committed to independent scholarship and supporting only as much infrastructure ~s absolutely necessary. Herein lies the root of a dilemma for Hopkins: in the Information Age, it is no longer possible to advance research and promote excellence in individual scholarship without an adequate information technology infrastructure. This, by it's very nature, requires common standards and goals, which are best attained through coordinated, cooperative effort. Recognizing the need to respond to this and other challenges, in 1992 Hopkins established a Committee for the 21st Century (C21) to "examine critically and imaginatively every aspect of the University's organization and programs...". C21 members were charged "to think along radical and fundamentally new lines ... to assume that many current arrangements will be outmoded and unsustainable by the end of the 1990s...". In other words, if Hopkins is to prosper and remain in a leadership role then nothing can be set aside as sacred, not even our long standing decentralization. C21 also recognized the critical role that information technology would play in any selftransformation, stating that "Universities that succeed in exploiting these technologies in a cost-effective manner will flourish; those that cannot will diminish in stature." PROJECT ORIGINS Despite these laudable efforts, the fact remains that Hopkins could ill afford to wait until the 21st century to establish basic institutional networked information resources. At many schools the establishment of a CWIS was officially mandated by the administration; at others it was initiated by major information providers; in most cases it is operated by or through the computing center. Given that none of these were probable mechanisms at Hopkins (there is no CIO or equivalent position, no university computing center, and even Communications and Public Affairs is decentralized), the question becomes: in an institution focused on individual achievement and where the dominant perspective is that of each separate division, how does a common information system come about? Certainly it would not spring into existence by spontaneous generation. In a sense, however, that is actually what happened, and to some extent this validates Gilman's essential thesis: individuals striving for excellence in pursuit of their goals laid the groundwork for what later evolved into JHuniverse, ultimately advancing the entire institution. The subsequent evolution of the project also reaffirms the value of Hopkins' lack of bureaucracy and willingness to encourage its faculty and staff to challenge the status quo and to implement new ideas and approaches. Early Adopters In 1991, faculty and staff members in several areas of the university were investigating the use of newly available Internet applications such as Gopher and WAIS as tools to enable access to research and archival information and to support collaborative work in their disciplines. These included, among others: Prof. Robert Kargon and his son Jeremy in History of Science; Special Collections and Archives of the Milton S. Eisenhower Library, primarily the efforts of staff member Brian Harrington; and the groundbreaking work of the JHU School of Medicine's Welch Medical Library with the Online Mendelian Inheritance in Man and the Genome Data Base projects, and the related Computational Biology. In October of that year the current technical manager of JHuniverse heard a presentation on CWISes at EDUCOM, and as a result he then attended in the spring of 1992 the American Society for Information Science Mid-Year meeting on networked information systems. This proved to be an revelatory event, occurring as it did at a significant juncture in the development of this field. The conclusion seemed obvious: the combination of ever more powerful computers, a robust and ubiquitous network, and simple, reliable information dissemination and navigational tools would lead to a revolution in the way computers are used and to an immeasurable increase in their value for research and scholarship. The ASIS meeting made it clear that these developments were more than simply another, albeit very effective, communications medium, and they could potentially provide much more for Hopkins than just a typical CWIS (and since JHU has several campuses, the term was never really apropos in the first place). Here, finally, was the realization of the promise of information technology to provide a solution to a problem that is otherwise effectively insoluble-to transcend the bounds of physical and political geography by providing a "virtual commons", a shared information and community space for an institution perpetually in danger of succumbing to the centrifugal forces which threaten to decompose it into a ''multiversity''[l] connected only by a common name; or worse, to tear it apart altogether. It is this idea that was the inspiration for the first part of the title of this talk. Information technology is one of the few tools that could provide a way for a decentralized institution to have its cake and eat it too-that is, sharing information, collaborating more easily, and participating in a common communication environment, while remaining dedicated to the simple principle of individual excellence. Pilot Project As it turned out, the Unix system administrator for Academic Computing at the JHU Homewood campus had already ported the gopher server software to run on our Unix system. This was quite a pleasant surprise; that there were also several fully developed gopher-based information services already operating at Hopkins was downright exciting. An interesting challenge was how to get them to join together under one "root gopher" when each viewed itself as a separate research project with no particular interest in the larger picture. We chose not press the issue, but instead tried to make it to their advantage to cooperate. As the official keepers of the jhu.edu domain, Homewood Academic Computing established the address 'gopher.jhu.edu' and registered it with the Univ. of Minn. as the main point of contact for The Johns Hopkins University. We publicized the availability of gopher, wrote articles about the wonders of gopherspace, distributed pre- configured versions of gopher client software and offered free short courses on Internet topics. By supporting the efforts of the early adopters and making it easier for Internet users to find their services, we engendered a good degree of cooperation and over the next year brought on board many new and enthusiastic participants. Victims of Our Own Success By the spring of 1993, Homewood Academic Computing had established a modest but successful network-based information service. This was effectively a pilot project, even though we did not initially conceive of it as such. However, it quickly became clear that continued development of the project would not proceed without the involvement and support of other pertinent groups outside of computing. Homewood Academic Computing has a limited mission within Hopkins, and its resources are insufficient to support a full-blown CWIS. We were fortunate, however, that the project was well positioned to play into a series of converging events which were about to catapult it into a new phase. GENESIS OF THE COLLABORATION In the summer of 1993, talk of the "Information Superhighway" was just beginning to explode into the national spotlight. The significance of this was not lost on the leaders of three major Hopkins service organizations, who separately were investigating the potential application to their organizations and looking for ways to capitalize on the networked information revolution. These were Larry Benedict, Dean of Student Affairs; Scott Bennett, Director of the Eisenhower Library (now Director of the Yale University Libraries); and Dennis O'Shea, Director of News and Information. In various discussions with each other they recognized their common interest in and need for a campus- wide information system. David Binko, Director of Academic Computing, brought to their attention the prototype CWIS that already existed in his department, and offered this as the basis on which to build a complete system. He also noted that there were no significant technical barriers in offering such a service-all that was needed was a sound organizational basis for a widely-used CWIS. Defining the Rationale for Action Three of the units sponsoring the new CWIS carry on extensive service programs and need to disseminate substantial amounts of information about these services. The primary activity of the fourth unit-News and Information-is the dissemination of information university-wide. It was clear that these four organizations would carry out their missions more effectively when their clients have access to a highly flexible, technologically advanced, and easily used CWIS. They all produce numerous publications and other documents which could be made available via the CWIS, so that communication with clients would be more timely, accurate, and potentially more costeffective. Each understood the tremendous advantages that might be realized by leveraging the substantial investment in networking infrastructure and desktop computers, which at that point was being vastly underutilized. What is notable about this is not that the four organizations recognized their common interests and the value of working together, as this was fairly obvious, but rather that they developed a pragmatic action plan and implemented it almost immediately. This group held its first meeting in September, formalized the details of the collaboration in October, announced the project publicly in December, and went on-line in January. This kind of rapid response was to become characteristic of the project. Hopkins has remarkably few barriers to impede innovation, which is a fundamental strength of the institution. In addition, there was a marked lack of territoriality among the four sponsoring units, which is unusual to say the least. Setting Fundamental Objectives In order to maximize the effectiveness of the system, certain defining characteristics were agreed upon which have proved to be critical to the subsequent success of the project: * No initial cost for information providers or consumers * Universal access (dial-in, telnet, etc.; no login required) * Flexibility (central coordination but local control) * Broadest possible audience (internal, external, alumni, etc.) * Simplicity (easy to use and to implement) While some of these objectives were the subject of considerable debate, they have resulted in a system which is uniquely attuned to the nature of our institution and which is truly inclusive. Some were agreed upon largely because we had no alternatives, but have proven to be valuable in their own right. For instance, we cannot load information for other departments because we do not have the staff to do so. Instead, we rely almost entirely on the departments themselves to upload their data. They retain control over their information and of the directories where it is located. For our part we provide written guidelines and instructions, software tools, classes and personalized training. This system borne of necessity has lead to a genuine sense of ownership, and a high level of commitment to the project. By committing ourselves to serving the university community in the broadest sense, we hoped to become a primary means by which the community informed itself, and in so doing developed a greater sense of itself as a community. This objective would be of value to any institution, regardless of size, budget, or organizational structure. Distributing the Work, Capitalizing on Our Talents To collaborate means literally to work together. In our case, none of the founding units could have successfully carried off this project on its own, so working together was a first a simple necessity. As a whole, however, we proved to be much greater than the sum of our separate capabilities. Initial organizational decisions were primarily based on the best fit of the skills and resources of each department, with each bringing a unique and complementary set of strengths to the project. The Eisenhower Library provided critical early leadership, and lent the project scholarly credibility. The participation of Student Affairs gave the project credibility with the schools and the administration due to their institutionally-critical mission. Academic Computing had the computer and network resources, technical expertise, and training skills to ensure that these vital aspects of the project were handled professionally. News and Information brought to the project their extensive communication and presentation expertise, organizational skill, and an important public relations perspective. Thus, from the very beginning JHuniverse has given nearly equal consideration to serving the local Hopkins community and to providing information about Hopkins to our extended family and to the world at large. The ongoing leadership role of News and Information (a part of university Communications and Public Affairs) has proven to be instrumental to this project. Robin Suits of Wright State University argues in a recent paper ("Campus-Wide Information Systems: A way to leverage information technology investments to meet strategic communication goals", obtained from the author, rsuits@nova.wright.edu) that a communications office may often be best suited to running a CWIS, since it is in their charter to "focus on the big picture of cross-disciplinary communication" and because "it also is the office charged with employing the most effective tools and techniques for meeting a university's strategic communication needs". In this respect the JHuniverse team is nearly unique, and has developed a management approach which is well-suited to the evolving nature and increasing importance of these systems to the institutions they serve. SECURING INSTITUTIONAL SUPPORT Building a successful Campus-Wide Information System requires developing the support of those who are going to use it, presumably everyone on campus--faculty, students, and staff-- as well as alumni, prospective students, other academics and the general public. Because some people at Hopkins had already created their own gophers at Hopkins and due to the decentralized nature of a university in which most people are unaccustomed to centrally coordinated systems, it was doubly important to develop strong support for or, at least, acceptance of our system. We began by building formal support from the units that provided the start-up funds for JHuniverse, and informal support from the broader university community. The four founding units agreed from the beginning that News & Information and Homewood Academic Computing would administer the CWIS. Those departments together with the other two founding units that also formed a Policy Board that would oversee the project. Initial funding for the project consisted of donations from each department based on the size of their budget, along with donations of staff time and other in-kind services. These commitments were guaranteed through FY96. One-time startup costs (mostly equipment) were evenly divided between the four units. The project manager developed a plan detailing the responsibilities of each of the managers and the policy board. Although we have not had much need to refer back to this document, it helped clarify our respective roles and responsibilities in project, and also brought to light several philosophical issues about the goals and mission of JHuniverse that were important to discuss. Informal Support The only group whose formal support the managers sought in the beginning was that of the JHuniverse Policy Board. It is important to note that while the Policy Board meet four or five times between the inception of the project in September 1993 and when it went on-line last January, the Board did not become involved in the day-to-day management decisions of JHuniverse. This hands-off approach allowed the managers to develop the system quickly and avoided any of the negative effects that committees can, at times, have on projects. Other than Policy Board, the managers did not seek formal support from the university. We did, however, inform the Provost that we were undertaking this project and requested $5,000 to help fund the networking of the News and Information Office. The request served to gain additional financial support and to inform the Provost of our activities. The Provost, in turn, urged us to make sure that the Policy Board included individuals from throughout the university since the original members were all located on one campus. In response, we expanded the board to include the director of the Welch Medical Library who was also the head of the new Biomedical Information Sciences division on the East Baltimore campus and an individual from the School of Continuing Studies' Montgomery County campus. We spent a great deal of time soliciting additional informal support from the university community. First, we publicized our efforts in the university-wide newspaper of whichthe project manager, conveniently, was editor. We also called and talked regularly with key faculty members, computer administrators, divisional administrators, other staff members and students. We also talked informally with senior officials in central administration. Another form of dialogue occurred via e-mail. Surprisingly, the largest volume of mail came from alumni who were extraordinarily interested in our efforts and wanted to see if we could provide additional services such as e-mail. The project managers personally and promptly answered every e-mail letter. We also conducted a great deal of outreach, targeting departments and offices that we felt should put information on-line sooner rather than later, and in the process we discussed the project with them and solicited what types of services and information they thought should be included on JHuniverse. Some of these meetings were one-shot deals, others involved a series of meetings over a period of months. Early on in the process we felt that we should create an advisory board made up of faculty, staff and students to act as brain-storming group. Because we have been so short staffed and struggling to keep up with the most pressing issues, we have not created such a board yet although we will before the end of the year. Their advice will be especially important as we design a comprehensive WWW-based system. Formal Support This fall, the project team recognized the need for dedicated staff. The initial funding provided by the consortium is simply not sufficient to even respond to current demands of the gopher-based CWIS, much less the more complex demands of a multi-media WWW-based system. We therefore decided to approach the university for additional funding. Accordingly, we discussed with Ross Jones, vice president of the Project Manager's division, our progress on the project and the need for additional funding. That meeting ultimately resulted in a presentation the Provost's Information Systems Coordinating Council, a university-wide group of senior administrators responsible for technology- related issues. A copy of our written presentation was also given to Hopkins' president, William C. Richardson. While no commitment has been made yet about providing additional funds for the project through FY94, the response from everyone has so far been very positive. One of the advantages of having a system in place and developing informal universitywide support has been that when we did formally present the concept to the central administration, we had a strong track record. One point which has generated a great deal of interest in these presentations is the fact that four major grants awarded to Hopkins were tied, to some degree, to the fact JHuniverse existed. A fifth extremely large grant was received, in part, because the information and expertise they develop will be made available on the Internet through JHuniverse. We have been able to show that not only is JHuniverse necessary to survive in the last few years of the 20th century, but that it also makes economic sense, generating research grant revenue and saving money in publications and marketing for undergraduate admissions. Tapping into Emerging Institutional Priorities The need for senior institutional support is critical at this point for JHuniverse for more than financial reasons. This fall the university released a report by the Committee for the 21st Century. A report issued by a subcommittee of C-21, states: "...the university must be committed to establishing an advanced managed interdivisional information system for students, faculty, and staff to generate, manipulate, preserve, and communicate information of all types in carrying out their education, scholarly, administrative activities." We needed to point out to the university administration that JHuniverse was such a system, and to ensure that other duplicative projects were not created due to a lack of knowledge about JHuniverse. We also needed to become more formally involved with others at Hopkins working on information technology projects. Finally, we needed top level university support to resolve interdivisional issues. The creation of centrally coordinated information systems often highlight of the problems that can develop within highly decentralized institutions. For example, the Johns Hopkins Institutions have at least three phone books, with different and, at times, conflicting information. There was a limit to the authority that the project managers could exert to resolve these types of problems. In our presentation to the Provost, he agreed, noting that it was time for us to "come in from the cold." Future Plans for Increasing Support As mentioned earlier, we will be forming a university-wide committee of faculty, staff and students to serve as a source of ideas for ways in which we can expand and improve JHuniverse in both the Gopher and WWW formats. Another top priority, if we are able to expand our staff, will be expanded outreach. We need to contact and work with many more departments and offices throughout the university to help teach them how they can use JHuniverse to reach their academic, research and organizational goals. We will also continue to provide training for information providers so that they can continue to upload their own information as we move into WWW. The Policy Board will undoubtedly be further reconstituted in the future and may become a subcommittee of some other existing university body, such as the abovementioned Council. Additionally, we need to work more closely with other Hopkins' groups that are working on technology-related issues such as placing student registration on-line and distance learning, areas in which JHuniverse has yet not been involved. Turning Good Intentions Into Success In a cooperative project such as this that is going to be used by a wide spectrum of individuals, it is essential to be fair, open-minded, flexible, and accommodating. When we started this project, each member of the consortium had different needs and expectations. When the Policy Board met, it was important to explore these different philosophies. Also, we found that our vision for the project expanded rapidly. While some others in the group had not envisioned JHuniverse as being, for example, a research tool or a forum for alumni, they were willing, fortunately, to allow us to experiment. Another important issue that arose early on was the format of JHuniverse. One of the librarians on the Policy Board was concerned that it didn't follow an organizational pattern that might have been created by librarians. However, the former Director of the Eisenhower Library noted that while this might be true, the new format might better serve the purposes of the project. Throughout the process of developing JHuniverse-an ongoing process that may never be "completed"-we have felt that it is critical to listen and respond to both suggestions and criticisms. This is a system that everyone owns, and the more input and support we receive, the stronger it will be. In a sense we must treat our users with the same level of respect as we would customers of a commercial product, because if they do not like JHuniverse the technology exists for them simply to create their own system, and at Hopkins they can and will. This does not mean, however, that anyone within the university will be charged for using JHuniverse. The founding members of the consortium were adamant that it be a free system. While we initially planned to cover the cost of developing and administering the system by charging information providers, we quickly decided this was not practical. We continue to explore options for generating income for the project without imposing direct chargeback. As we approach our first anniversary and begin to "come in from the cold," the stakes are getting much higher. If we are more formally sanctioned and funded by the university, we will have to build a larger, more comprehensive system. University scrutiny will increase, and there is the danger that University funding will entail centralized control, with the consequent loss of some of the of the unique features of our current collaborative arrangement. Balancing these forces will be difficult, but we look forward to the continuing challenge of providing innovative and quality service. THE FUTURE IS NOW As information technology continues to advance and as JHuniverse grows, there will be many issues and problems that will arise. The key for us will be to maintain that delicate balance between stability and technological advancement. Thus, it is imperative to stay informed of new developments, to keep an eye on the future and always have something in place to anticipate it, but also to not change things so often that people get frustrated. Despite the lure of better software and new organizational ideas, for instance, we try not to significantly alter JHuniverse more often than twice a year. Even so, it is important to be willing to take risks when the reward is potentially great. Our greatest reservoir of creativity and initiative is our students, and by tapping into this source we can make great strides forward very quickly. This is why we recently made it possible for students to publish their own WWW "homepages" through JHuniverse, even though we have not worked through all the administrative issues just yet. We would rather seize the opportunity while the potential rewards are greatest, and deal with any unforeseen consequences later. We must continue to focus on content and communication, rather than on technology. If we are able to do this and to maintain the cooperative spirit that has seen us through this far, we will continue to meets the needs of the Johns Hopkins community. 1 Attributed to Milton S. Eisenhower, eighth president of Hopkins, in "A Brief History of Johns Hopkins University", John C. Schmidt, JHU Press, Baltimore, 1984.