Lessons From The Berkeley Museum Informatics Project Copyright CAUSE 1994. This paper was presented at the 1993 CAUSE Annual Conference held in San Diego, California, December 7-10, 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 technology 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; 303449-4430; e-mail info@cause.colorado.edu Lessons from the Berkeley Museum Informatics Project Barbara H. Morgan Director, Advanced Technology Planning University of California at Berkeley Berkeley, California December 8, 1993 barbm@garnet.berkeley.edu The Museum Informatics Project at the University of California in Berkeley is a collaborative effort to coordinate the application of information technology in the more than 80 museum and other organized, non-book collections on the campus. Faculty, collection managers, curators, librarians and information technology specialistswork together to develop data models, system architectures, demonstration systems and network-accessible production databases of text records, images, sound, video and film. This presentation examines some of the early successes and difficult problems of the project and its predecessor activities, to show how the necessary political, administrative, financial and technical support for the project was created within the institution. INTRODUCTION Why is the Berkeley Museum Informatics Project of interest? New alliances are needed so that we can share scarce resources internally on our campuses and leverage external resources. The Berkeley Museum Informatics Project is an example of a new alliance on our campus, and there may be lessons from the successes and problems of this collaborative project and its predecessor activities that apply to other cross-organizational and cross-disciplinary efforts. This examination will show how the necessary political, administrative, financial and technical support for this project was created within the institution, and what kind of leadership was needed to initiate a change effort of this magnitude. Audiences Audiences who might be interested in this topic, in addition to individuals embarked on similar efforts at other institutions, include university and college administrators who are responsible for many disparate units and who see promise in information technology for better management of their operations. In particular, people contemplating large-scale ventures to implement imaging systems for paper documents might find useful parallels with this Museum Informatics Project. The project may also be of interest to information technology professionals who are concerned about the problems of moving from fancy multimedia prototypes to large-scale production systems. University of California at Berkeley The University of California at Berkeley is part of the nine- campus University of California system, a public institution. We have 31,000 students (22,000 under-graduates and 9,000 graduate students) from 100 countries. Three hundred degree programs are offered in 14 colleges and schools. We have a law school and a business school; we have no medical school. There are probably about 17,000 computers on campus; about 12,500 are connected to the campus network. MUSEUM INFORMATICS PROJECT Summary The Museum Informatics Project (MIP) is a collaborative effort to coordinate the application of information technology in the museum and other organized, non-book collections on the campus. The Berkeley campus has more than 80 collections, with 30 million artifacts, housed in 65 organizational units in all major academic disciplines. A few examples of the many kinds of materials in these collections include costumes, bugs, forest products, kinetic art, Egyptian mummies, plant specimens, soil samples, fossils, maps, architectural drawings, temple rubbings and musical instruments. The project was formally created in January 1992. The founding members were: Jepson and University Herbaria Museum of Paleontology Museum of Vertebrate Zoology Phoebe Apperson Hearst Museum of Anthropology University Art Museum Architecture Slide Library History of Art Visual Resources Collection Information Systems & Technology MIP's goal is to broaden access to fundamental academic resources by scholars, students and the public. Participants in the project include faculty, collection managers, curators, librarians and information technology specialists. MIP has collaborative relationships with other University of California campuses, especially those at Davis and Riverside, and with the University of California Office of the President. Berkeley participants are also collaborating with people working on similar activities at Harvard, Cornell, MIT, USC, the University of Washington and the Australian National Botanical Gardens, among others. Activities and architecture The Museum Informatics Project has these major activities: + Produce a campus strategic plan for museum informatics; + Develop guidelines for information systems in museums and other collections; + Maintain an information clearinghouse; + Operate a demonstration and development facility; + Evaluate, select and implement information tools for scholars and curators; + Develop data models, system architectures, demonstration systems; + Develop network-accessible production databases of text records, images, sounds, video and film; + Assist museums migrating from legacy systems; + Obtain extramural funding for campus museum informatics efforts; + Provide a forum for serious discussion of intellectual property issues and other matters related to scholarly databases; + Produce a comprehensive catalog of campus collections. The project has a strategic architecture (shown in the diagram at the end of this paper). The network is the key element. The campus high-speed network connects every collection and every user, and the Internet connects the collections to other scholarly institutions, public agencies, private non-profit organizations and the general public. Collections are grouped intellectually in terms of their relationship to biological diversity, cultural diversity, and physical diversity. Client-server architecture and distributed systems imply that each function is carried out where it can be done best and is most needed. Each resource is managed by the parties most immediately responsible, but access, including use of shared services and systems, can be granted from any location. Strategic support is provided by the Academic Senate, the campus administration, the central computing organization, and the Library. "Cooperative autonomy" is enhanced by planning and guidance from advisory committees and functional working groups. Management and funding Administratively, the Museum Informatics Project is a department of Information Systems & Technology (IST), the central computing organization. It is directed by a faculty member, Professor Thomas Duncan, who also has an appointment in the Department of Integrative Biology. The project has a Scholarly Uses Advisory Committee (composed primarily of faculty interested in research and instructional uses of the data in various collections), and a Collections Management Advisory Committee (primarily for curators and collection managers). Functional Working Groups have also been created in 11 different areas, to work on technical, operational and policy issues. Examples include: Imaging and electronic multimedia Geographic information systems Shared authority and reference files Intellectual property rights The Museum Informatics Project core staff (six FTE), and its Development and Demonstration Lab, receive State funding as a unit of Information Systems & Technology. Staff from other units of IST who are participating in the project, as well as staff and faculty from the departments housing the collections, are also paid by various campus funding sources. The National Science Foundation is providing substantial support for one of the MIP projects; the U.S. Department of Education has also provided some funding. Several grant proposals to State and Federal agencies are pending. Many hardware and software companies have supported the project; additional corporate support is being pursued. The Museum Informatics Project has been one of many campus groups involved in planning for a major new fundraising campaign for Berkeley. HISTORY AND LESSONS At this point in discussing the Berkeley activities I would normally describe some of our current projects, show some examples of materials from our electronic collections, and comment on our successes and difficulties, with suggestions about pitfalls that other groups might avoid. However, for this CAUSE Conference, I have been thinking about how this ambitious effort came about and what lessons there might be for similar efforts at other universities. Preliminary work actually began in 1987, almost five years before the Museum Informatics Project was formally announced to the campus. Staff members working with three collections, the Architecture Slide Library (Maryly Snow), the Geography Map Library (Daniel Holmes) and the University Art Museum (Howard Besser), came to the chief campus computing officer (Raymond Neff) for help in creating "image databases" of their materials. Dr. Neff agreed to fund a collaborative pilot project, but only if the images were digital and the collections could be accessible on the campus network. A group I direct, called Advanced Technology Planning, was formed in 1988 and became responsible for image- database software development activities. I am not going to recount the interesting events that have happened in the last six years. Instead, I want to talk about factors that came together to make it possible for the campus to create this large-scale, cross-disciplinary, cross-organizational, multi-year effort to provide access to the resources in our museum and other organized object collections. It seems to me that at least eight elements of support had to be present to create the conditions for the Museum Informatics Project. This complexity turned out to be a strength; different entities were active at different times so that, when support weakened in one area, it increased in another. This project and its predecessor activities show the synergy that can come about from administrative commitment, government support, commercial support and individuals who believe in what they are doing and are willing to work very hard. Sustained high-level administrative interest The single most important factor supporting the Museum Informatics Project has been the personal interest and consistent support by our Provost for Research, Joseph Cerny. Information Systems & Technology has been reporting to Dr. Cerny since 1990, but his interest pre-dates that arrangement. He has made additional resources available to other campus units besides IST, most notably the Hearst Museum of Anthropology, and he has been a strong advocate of museum informatics since before we knew what to call it. The central computing organization has made this cluster of activities a priority for more than six years, through four different information technology management arrangements (three different heads of computing, plus a faculty committee). Our current Vice Provost for Information Systems & Technology, John McCredie, is a strong supporter of the Museum Informatics Project, and it reports directly to him. The right technology We needed enough network connectivity and bandwidth to make both the earlier image database and the current museum informatics activities possible. We needed software portability (provided primarily by Unix, by the X Window System and by SQL database software). The cost of equipment has been coming down dramatically, of course (especially digitizers, workstations, and storage devices). We are now seeing interesting new developments in cost-effective CD-ROM tools that are likely to make it possible to do both network-accessible databases and subsets on CD-ROM for special purposes or special audiences. Another aspect of "right technology" has been important: there is now enough technology in place on our campus, as a normal part of everyone's daily work, so that it is easier for people to think of computers as being helpful, to imagine sharing ideas and data, and to believe that disparate groups might have similar "information" problems. Image databases are attractive applications, and the prototype databases we developed have had broad appeal: many people who don't think much of computers or networking can imagine the possibilities. Good programmers We had an excellent programming team working on the original prototypes, with a particularly brilliant and productive team leader (Steve Jacobson, now with Franz Inc.). We have been able to continue a strong programming team (currently led by Randy Ballew) through several personnel changes. In addition to prototypes, the Advanced Technology Planning group has been designing and implementing production databases for some of the collections. We are in a slow process of turning those databases over to programmers working for the Museum Informatics Project and to programmers paid by grant or departmental funds to work with particular collections. Faculty leadership In spite of successful prototypes, the Museum Informatics Project would not have become a major campus activity without the leadership of a dedicated faculty member (Thomas Duncan), who was willing to make enormous personal and professional commitments to the project. We had difficulties in the early years of the image database work because we did not have faculty "champions" for most of the prototype projects. Our small teams of curatorial staff and computer center programmers were sometimes in awkward positions relative to the priorities in academic departments, with problems obtaining resources from deans and provosts, or from external funding agencies. Professor Duncan has been much more effective than staff had been in persuading other faculty to participate in strategic planning, grant writing and data modeling activities. Increasing faculty involvement continues to be a high priority for the Museum Informatics Project. Vendor interest We have had support from many hardware and software companies since the earliest days of the image database project. Support has meant much more than just encouragement: we have received donations and loans of hardware and software, been sponsored at conferences, and had frequent opportunities to talk with corporate software development people. Early supporters were Image Understanding Systems, Ingres, Sun Microsystems, Apple Computer and IBM. Modest support was also received for the image database work from Digital Equipment, MIPS, NeXT, Oracle and Pacific Bell. At one point Carlyle Systems licensed the Berkeley image query software, under an agreement involving the Museum of Anthropology collection; although that arrangement did not work out, the experience was instructive for all concerned. Sun Microsystems has been our most consistent supporter. A large donation to the fledgling Museum Informatics Project in early 1992 helped put equipment into eight different departments on campus. This donation proved to the potential MIP campus participants that collaborating can have tangible results, and the timing was helpful in obtaining National Science Foundation support for one of the MIP projects. Discussions are currently underway between MIP participants and several companies, mostly notably Apple Computer, Eastman Kodak and Silicon Graphics. Government support The National Science Foundation (Biological Research Resources Division) has provided substantial, multi-year funding to the Specimen Management System for California Herbaria (SMASCH). We believe that the campus commitments to the Museum Informatics Project and to collaboration with the national (and international) botanical community were major factors in NSF's decision to make this award. The U.S. Department of Education (College Library Technology and Cooperation Grants Program) was our first Federal supporter, with funding in 1991 for a 40-gigabyte image server for the Architecture Slide Library collection. Proposals have been submitted recently by Museum Project participants to the National Endowment for the Humanities, the National Endowment for the Arts, the National Science Foundation and the California State Library. The next two years should see a substantial increase in grant-writing activities. People who own the data are willing to work together The original request for computer center help in 1987 came from campus individuals who cared about the data in their collections and who were willing to work together to try to create something mutually beneficial. The belief that campus groups need to come together to try and leverage scarce resources has been especially strong among staff curators and librarians for special collections. Recently we have seen increasing acceptance from both faculty and staff of the idea that collections in different disciplines may have common information management problems and opportunities. In addition, many interesting new relationships have been established among groups and individuals who had never worked together in the past. In 1989, 25 individuals from 12 campus organizations participated in an evaluation of collection management software. Eight organizations founded the Museum Informatics Project; 18 were represented at the first formal meeting in the winter of 1992. There was an excellent response to MIP's call for expressions of interest in early 1992. Thirty-six campus groups responded within two months, and almost 80 collections are represented in the extensive survey of the campus collections that MIP has been working on for the past year. Many individuals are participating in the Functional Working Groups, which began meeting in the fall of 1993. Many of the collections participating in the Museum Informatics Project are part of the campus Library organization. Library management has been supportive of the image database and museum software evaluation activities since 1988. In 1992, representatives from the MIP core staff, the Advanced Technology Planning group, and the Library began meeting monthly to talk about supporting "scholarly databases" and to make sure that each group's activities are coordinated and difficult development work is not duplicated. Visibility Berkeley people (both information technology professionals and the people with the data) have been willing to make presentations, demonstrate software, help other groups with demos, and write about the work we are doing. Most of us always say yes to opportunities to present in academic settings, and we will present in other settings if there is a potential benefit to the campus. Descriptions of the Berkeley activities by software and hardware companies which publicized our work have also been helpful. Visibility helped legitimize the Museum Project on campus, and it also brought us external collaborators: people engaged in similar activities and those who are interested in sharing data as well as design ideas. OPPORTUNITIES FOR THE FUTURE Now that the Museum Informatics Project is underway, where are we going? We are assuming that fast, high-capacity, easy-to-use computing technology will become available all over the world in the next decade. Technology will become affordable for most institutions and scholars, as well as for others who wish to access data from museums and other collections. By providing access to the data, images and sounds contained in our collections, we make the collections themselves more valuable, we get to use them more intensively, and we can also do a better job of conserving the artifacts and specimens. We can "liberate centuries of captive information," in the words of one of our participants: we finally have an opportunity to access information that traditionally has been ignored and inaccessible to many scholars, teachers and students. We can unlock the wealth of our collections. Most collections are underutilized. Many cannot be used for multiple kinds of study because of the extreme cost of finding pertinent materials. Museum informatics technology provides the way to exploit our campus intellectual resources more thoroughly.