The combination of digital imaging, distributed computing, and high-speed networking technologies has ushered in a new era in the use of image and information resources. Electronic communication is altering patterns of scholarship and collaboration. During the next decade, digital technology will play a vital role in enabling dramatic changes in university teaching and research. These changes will be accompanied by and promoted by transforming shifts in the organization, economics, and methods of higher education.
My comments focus on the development of digital collections based on cultural and scientific resources, and I will describe representative examples, analyzing the concepts underlying their composition and use. I will discuss the methods of funding and creating each and offer some critical observations on their success in meeting envisioned goals. The provision of networked access to research collections can be seen as a paradigm illustrative of more extensive changes affecting higher education and cultural repositories.
At Cornell University, a remarkable array of historical, artistic, and scientific resources are available via the Web. In my roles as Director of the University Library's Division of Rare and Manuscript Collections and of the campus-wide Digital Access Coalition, I have participated in the creation of several of these resources. The use of World Wide Web technology provided a critical transition from our initial CD-ROM applications. In the case of some users and uses, the change was not initially viewed as an improvement, but the Web has quickly become the standard for networked access to digital collections. Although the underlying technologies may change significantly, it is likely that the Web will remain the model and the metaphor for ubiquitous, global, Internet computing for some time.
In using the term digital collection, I will be primarily referring to digital surrogates of images, objects, sounds, specimens, and texts that presently exist or have previously existed in other media, but there is no reason to exclude items that have only existed in digital form from this definition. Collections are usually based on common thematic characteristics and/or common provenance of creation or organization, and are shaped by curatorial intent, projected use, and issues of ownership and project delineation. Collection definition is also provided by the links employed in the organizing of access to and navigation within the collection.
On-line exhibitions are not the same as digital collections. In most cases, curatorial intent, projected use, and selection criteria differ significantly. It is likely, however, that the two will often be employed in combination.
The principal Cornell digital collections (rmc.library.cornell.edu) considered here include:
Each of these projects represent different organizational and professional collaborations. At Cornell, the principal participants have been the Division of Rare and Manuscript Collections (RMC) , the Digital Access Coalition (DAC), and the Interactive Multimedia Group (IMG).
RMC is Cornell's major repository of antiquarian printed books, manuscrip ts, and University Archives. It is also the largest repository of visual resources, holding over one million photographs, paintings, prints, and other visual media.
The Digital Access Coalition is a cooperative effort by librarians, archivists, curators, technologists, and teaching faculty from across the campus to develop a new vision for organizing, accessing, and using the University's various historical, ethnographic, artistic, and natural history collections. The Coalition was established in 1992 with the support of the University Librarian and the Vice President for Information Technologies. It is located in the Library and operates under the auspices of the University Provost. It was co-directed by Geri Gay (Associate Professor, Department of Communication) and myself from 1993 through 1996.
The Interactive Multimedia Group was formed in 1987 by Professor Gay. IMG staff regularly engage in the design, research, and teaching of interactive multimedia computing. Evaluators, designers, programmers, and researchers work in concert to develop interactive multimedia programs that meet the needs of users, demonstrate key issues in multimedia design, and result in research findings that contribute knowledge about leading-edge interactive technologies.
Other campus partners include the Herbert F. Johnson Museum of Art, the Department of Preservation (University Library), the Laboratory of Ornithology, the History of Art Department, the Theory Center (Center for Theory and Simulation in Science and Engineering), the slide libraries of the College of Architecture, Art, and Planning and of the History of Art Department, the Computer Science Department, Cornell Plantations, and Cornell Information Technologies.
The diverse alliances evident in these projects are a central aspect of this new technical and organizational environment. Realizing the full potential of emerging technologies to both improve access to university collections and to enhance research and teaching requires the development of new cooperative models that are dependent on interaction across professions, as well as disciplines.
Louis Agassiz Fuertes was an accomplished naturalist recognized by ornithologists as this country's greatest painter of birds. Fuertes' paintings are held by museums across the country, but the largest collection of his artwork, including over 1,000 pieces, is owned by the Cornell Library. In addition to his paintings and drawings, our collection also includes letters, sketchbooks, diaries, field notes, and photographs. At Cornell, Fuertes' works are also held by the Johnson Museum of Art and the Laboratory of Ornithology.
The Fuertes site includes a "virtual expedition" tracing the 1899 Harriman Expedition to Alaska through the text of the journal that Fuertes kept during the trip. This section is designed to provide access to information about the expedition in a variety of ways, including text, maps, images, and sound. An initial screen shows a map tracing the route of the expedition. Clicking on any spot along the route will retrieve the journal pages describing events at that location. Journal pages can also be retrieved sequentially, or by specific date. Within each journal entry, the names of birds are highlighted. Clicking on the highlighted text will provide thumbnail images of that bird, as depicted by Fuertes. Attached to each thumbnail image are links to the full-screen image, information describing the painting, and to the bird's call or song, based on digital recordings from Cornell's Library of Natural Sounds. Journal pages have been transcribed for easy reading, but bit-mapped scans of the original manuscript pages are also linked to each transcription. These images provide a researcher with the original handwritten text, but are also critical for the drawings they contain and his visual renderings of bird songs.
In its original version, the site also included an electronic version of Of a Feather: Audubon and Fuertes, an exhibition at Cornell's Johnson Museum, July 5-August 21, 1994. This "exhibition without walls" allowed the user to view the exhibit by clicking on a particular "gallery" on the screen, displaying thumbnail images of the pictures exhibited in the Museum. Clicking on the image itself provided a full-screen version of the image. Additionally, there were comparisons of the artwork of Alexander Wilson, John James Audubon, and Fuertes, as well as links to the Alaska Expedition journal. The site is visually and intellectually engaging and was recognized by PC/Computing as one of 1995's "best free sites on the Internet."
The initial conversion of Fuertes' artwork was conducted as a component of Cornell's participation in the Kodak Library Image Consortium. This joint project of Cornell, the University of Southern California, the Commission on Preservation and Access, and Eastman Kodak Company was designed to test and evaluate the applicability of Kodak's Photo-CD technology for the kind of large-scale use envisioned by museums, libraries, and archives (1992-1994). The development of the Web site was funded in part by the Council on Library Resources as part of a project to develop and test tools for multimedia access to collections (1994-1995). Today, we have withdrawn the exhibition section of the site and have added Web access to 2,648 images with complete cataloging and indexing for each image. These works represent the complete holdings of all three Cornell collections. Completing image conversion and entering existing cataloging data was supported by a private gift. Although the creation and maintenance of this site has been partially funded by external resources, significant costs have still fallen on the Library. A rough approximation of total project direct costs is $55,000.
This was our first Web site, initially constructed in the summer of 1994, and I think that it is probably our most successful effort to date to provide a research collection digitally. We started out with a coherent theme that effectively integrates intellectual and navigational approaches. We envisioned five potential research constituencies: historians of science, art historians, biologists, ethnographers, and birders. The nature of our sources and the focus on the Harriman Expedition allowed us to think explicitly about the needs of potential users and what would increase the value of the site to each clientele. Importantly, we have also created a "virtual collection," bringing together works, that while being located on the same campus, have probably never been used in combination.
Other than multimedia design classes, this database has not been used systematically on campus so our evaluation data is selective. We have been dependent on observations by Cornell ornithologists and occasional comments from remote users. We have responded to these and have improved the site. We made a major addition by adding the complete collection, previously on CD-ROM, to the site. We also removed the exhibition. We did not believe that segment was sufficiently strong, and that there are real questions regarding the extent to which on-line exhibitions should be "permanent." On the other hand, research collections must be citeable over time. This suggests that one must have a clear sense of purpose and plan accordingly.
Our use of bird songs is amateurish. We have a good deal more to learn about the use of sound. Further significant enhancement could be made by adding the letters that Fuertes wrote home during the expedition. We have adopted a questionable practice in that our links to sketches and paintings for each bird include works Fuertes did elsewhere, as well as those made in Alaska. Because the painting are linked to the catalog with full descriptive data, it is relatively easy for a user to make such distinctions, but perhaps we are being misleading. Ethnographers could benefit from the inclusion of more photographs. Interestingly, these small snapshots are much more effective as digitized surrogates than in their original form.
A principal reason for the success of this effort is the effective integration of compelling visuals with significant textual information. We must remember that the speed with which the Web was adopted is based on the remarkable improvement these graphic displays offered over gopher technologies. While the novelty has worn off, this feature has become an even more essential requirement for attracting casual, instructional, and research users.
Under the auspices of the Digital Access Coalition, the Utopia Project has created a thematically-related electronic database of descriptive information linked to digitized images. The database is nearing completion and will include 5,000 images focusing on the art and architecture of the Italian Renaissance. Of the initial 5,000 images, approximately 4,000 have been selected from the slide collections of the College of Architecture, Art, and Planning and of the Department of History of Art in the College of Arts and Sciences, with the remaining 1,000 from illustrations in printed books in the Library's rare book collection and works of art on paper in the Johnson Museum.
This pilot implementation is an enterprise involving colleges and other units across the university and drawing on a wide range of personnel and resources. The project was developed by a steering committee comprised of Claudia Lazzaro, Chair of the History of Art Department, Margaret Webster, Visual Resources Curator in the College of Architecture, Art, and Planning, and myself, and is being administered by DAC. Financial resources have been provided by the Deans of Arts and Sciences and Architecture, Art, and Planning, the University Provost, and the campus Humanities Council's Fund for Research in the Humanities. Direct contributions have totaled $97,500. Additionally, hardware and software and server maintenance have been provided by staff of the Cornell Theory Center, under the direction of Marcy Rosenkrantz, Associate Director of Supercomputing Technologies. While the database has run on a desktop machine, not a supercomputer, this involvement by the staff of a scientific and engineering facility in a humanities-based project is unusual at Cornell, and I suspect elsewhere.
This database has had significant classroom use and some research use. It was initially used in a seminar on Renaissance and Baroque Prints in Spring 1995, and since then in Vincent Mulcahy's large Introduction to Architecture classes in Fall 1995 and 1996, in a Renaissance Culture course, and now in Professor Mary Woods' History of Architecture course. Systematic evaluation has been conducted in all of the courses. My colleague Noni Korf Vidal (Curator for Visual and Electronic Resources) has prepared a paper for this conference in which she describes the results in detail. In general, student evaluations are overwhelmingly positive. They found Utopia easy to use on the Web. Most agreed that it was an intuitive system and that they needed little instruction in order to begin using the database. They liked having a Keyword field with all the Subject and Object Type terms and significant words from the Titles grouped together. They were very enthusiastic about the Browse capability that allows viewers to browse through just the images in a search result, as well as viewing them with titling or full cataloging, and they would like to see the Browse feature extended to other searchable fields (at themoment, it's limited to Keyword, Object Class, and sets specifically chosen for a particular class).
On the negative side, slow response time is a ubiquitous source of discontent. Nor do students like having to go through two screens of information in order to look at the high resolution image. They want more flexibility to move around in the database without having to look at the full data for each thumbnail as an intermediate step. They want to be able to save the thumbnails from one search and add to or refine their search. Naturally, they want to be able to zoom in and out and download and print good quality images for future study or to review on their own systems.
Faculty evaluations have also been positive. They have greater concern about color and resolution quality, and they want more sophisticated searching capabilities. Most importantly, faculty feel that it is crucial to add more images in order for the Utopia database to fully meet their teaching and study needs.
Selection of images for this database has been conducted in direct cooperation with teaching faculty. This has insured its utility. I think that this direct involvement in shaping content is essential to use and critical to campus funding. Furthermore, we have provided significant technical support for initial implementations. This assistance will remain important for some time, and it puts us in a good position to make effective evaluations of content and technology.
Many of the images converted did not present licensing issues, but some do. We contacted several slide producers and in a few cases the museums holding the original works from which the slides were produced. Some licenses have been negotiated, and others are under discussion. We have limited access by IP address in order to comply with contractual commitments.
Cataloging of images has drawn on existing data but has also required some original work. All data has required editing and fielding, and we have developed browse features based on the data, as well as exact searches. Use by faculty in both architecture and the history of art has resulted in cataloging differences based on the manner in which these two disciplines approach their sources. These differences continue in theory but often disappear in practice. The on-line environment offers ways to harmonize distinctions that are problematic in traditional slide collections. The inclusion of searchable descriptions is a critical asset to the digital file. With today's projection equipment, slides are still more effective for classroom display of images, but searchable cataloging provides students with a much more powerful and creative means of conducting their study and research. Faculty have been surprised by associations among images based on features other than the traditional elements of creator and location. This kind of unified, yet dynamic, access to both information and images has the potential to reshape teaching and research to take into account a wide range of cultural artifacts and to encourage interdisciplinary and cross-cultural approaches to learning.
Cornell is one of seven universities and seven collecting institutions participating in the Museum Educational Site Licensing Project (MESL). MESL links six museums and the Library of Congress with the seven universities in a collaborative research effort to define the terms and conditions for the educational use of digitized museum images and related information and to explore and promote the educational benefits of digital access to museum collections. The project was launched by the Getty Art History Information Program, now the Getty Information Institute, and will continue through June 1998. MESL participants are developing a model educational site license, testing and evaluating procedures for the collection and distribution of museums' digital images and information, assessing the impact of this distribution in both technical and economic terms, and evaluating educational benefits on the seven campuses.
During this project, the participating museums are providing digital images and information to the universities without charging any royalty fees. It is assumed that each institution will provide for campus-wide access and will make use of the images in unique ways. An assessment of the uses of the images at each university will provide information that will lead to a better understanding of how digital image providers and users can benefit mutually and provide the basis for future licensing arrangements.
The MESL database contains images of paintings, prints, photographs, textiles, and various cultural artifacts. The image selections from each museum vary greatly. For instance, the National Museum of American Art chose to present a sample of their collection by providing representative works from numerous artists. At the other end of the spectrum, the Library of Congress in its initial contribution chose to present the work of one American photographer, Carl Van Vechten. The objects themselves also vary. While the images from the National Gallery of Art are largely paintings, the Fowler Museum of Cultural History, the Harvard University Art Museums, and the Museum of Fine Arts in Houston provided a significant range of cultural artifacts, including textiles, furniture, ceremonial artifacts, and documentary photographs of objects in use. During the life of the project, originally two years and now extended to three, access to the images is provided freely to anyone at the participating institutions with the understanding that users will adhere to conditions of use designed to protect the copyrights of the museums.
At Cornell, faculty interested in using the MESL database were urged to contact the MESL coordinator at Cornell, Noni Korf Vidal, to discuss how we could help in incorporating the database into the curriculum, but we have had much less direct involvement in the use of the MESL database than in the case of the Utopia Project. In the case of Associate Professor Laura Meixner's use of images of paintings from the National Gallery of Art, the National Museum of American Art, and the Harvard University Art Museums into the teaching of her course, "Painting and Everyday Life in Nineteenth Century America," we created an on-line study guide. This year we have created a "Virtual Gallery of Impressionist Images" for use with her current course, "Impressionism and Society." Browsing the Gallery, one can explore major themes represented by French and American Impressionist painters. Also included are seventeenth- and eighteenth-century artists who significantly influenced the Impressionists. Professor Meixner and Johnson Museum staff have also curated an exhibit of Impressionist prints from the Johnson holdings and have installed a computer in the study gallery so that visitors can also access the MESL "Virtual Gallery."
Nationally, the overarching goal of the project is to define the terms and conditions under which digitized museum images and information can be distributed over campus networks for educational use. The three principal objectives are:
In pursuit of these objectives, numerous activities have been initiated. Formal studies are underway, including project economics (Howard Besser, University of California, Berkeley, supported by The Andrew W. Mellon Foundation) and a user study (Geri Gay and Robert Rieger, Cornell University, supported by The Getty Information Institute). The results of these studies and project reports are forthcoming.
Project collaboration has been impressive. These fourteen different institutions, working with minimal external support, have accomplished a great deal. The six museums and the Library of Congress each converted 1,000-1,500 pieces to digital form. (In most cases, conversion was conducted as part of this project, but a couple of museums had already initiated significant digitization efforts.) Image files, created by different means and using different conventions, were transferred to the University of Michigan, along with cataloging data for the images. At Michigan, some verification and normalization was conducted, and files were sent on to each university. At each university, access to the images and information was limited to faculty, students, and staff by IP address. Local mounting of the files and user access was conducted using different hardware, software, and approaches. Although at Cornell initial access was via CD-ROM, all institutions now employ Web access. Individualized approaches to conversion and independent mounting of the same database on seven campuses was wasteful, but it did result in the development of greater local experience and expertise than would have been the case with a consolidated approach.
Selection and distribution of objects was conducted in two rounds, including a minimum of 500 images each time. In the initial round, there was very little opportunity for expressions of faculty preference. Although the museums were responsive to requests during the second round, the timeline for conversion did not mesh well with the academic calendar. Greater faculty involvement would have been beneficial in insuring a better match between their interests and database content. Faculty are also hesitant to devote the necessary time to influence the development of a resource that may not be available in a year or two. I think that participants failed to recognize this issue in initial planning. Nonetheless, the database is very rich, and it has been used creatively across the country.
In concluding my comments regarding MESL, I want to express a few observations regarding the cataloging of museum holdings. In the initial compilation of museum data, the inconsistency between the cataloging from the various museums was surprising to us all, and significantly limited searching capabilities. This inconsistency was evident in terminology and vocabulary, but more importantly, comparable data elements were not always present. I do not think that this reflects a lack of common knowledge regarding cultural collections. I think that it results in part from the fact that existing collection management data was not been designed to provide user access. Without a tradition of public access to catalog information, description has been based on internal needs. Now a very different approach is needed, and I wonder if recent efforts to develop common descriptive standards have focused sufficiently on user access criteria. Existing museum computer systems also appear to focus almost exclusively on collection management processes. To some degree this is alleviated by the capabilities being designed into Web browsers and public kiosks, but future cataloging practices should be designed to support distributed research access.
Democratizing Access to Multimedia Collections is a privately-funded project directed by Geri Gay and myself. Principal objectives of this project included providing networked classroom and research access to manuscript collections, graphic arts, and historical artifacts and evaluating their use by college and K-12 students. The Web site, "Invention and Enterprise: Ezra Cornell, A Nineteenth-Century Life," is one of the products of this effort and includes: an on-line exhibition; a guide to the Ezra Cornell Papers in the Division of Rare and Manuscript Collections; 30,000 scanned manuscript pages; and links to books and journals digitized as part the Mellon Foundation-sponsored Making of America Project. In addition to broad general use, the Web site is currently being used by a local fourth grade class (Cayuga Heights Elementary School) as part of their course on New York State History.
The Ezra Cornell site is representative of a model that may become increasingly common. An exhibition with substantial visual and historical appeal provides the initial point of entry for a broad range of users, as well as providing an introductory narrative and introducing navigational devices. Some users may limit their use to this one component, but are encouraged through explicit links to access additional digital reproductions of documents, objects, and graphic images. Sites may also include archival finding aids, either providing an overview of the larger collection available at the holding repository or providing searching and navigational support for collection access on-line. This combination may attract a diverse audience while enhancing overall utility.
Another component in the Democratizing Access to Multimedia Collections Project was the creation of an on-line version of "Paper, Leather, Clay and Stone," an exhibition conveying a history of written communication over 4,500 years, concentrating on the way the medium of communication shaped and was shaped by the message. The original exhibition was created with a particular focus on a high school audience and was curated in cooperation with a local high school teacher and faculty from Cornell's Near Eastern Studies Department. Other project components include Professor Gay's collaboration with the British Museum in London, the National Museum of Ethnology in Osaka, Japan, and IBM-Japan in the creation of a Global Digital Museum (GDM). Museum staff are presently digitizing selected holdings from the two museums. These materials will form the core of the GDM Web site, and electronic learning resource packs (GDM Books) available to primary and secondary educators throughout the world.
In concluding this review, I will briefly describe three new projects. Over the next two years, the Herbert F. Johnson Museum of Art On-line Project will create high-resolution digital surrogates for the majority of the Johnson's permanent collection (20,000+ items), accompanied by description and indexing. Images of paintings, photographs, prints, and sculptures will be made available via the Web. On-line kiosks will enhance the onsite experience, incorporating the fullness of the holdings with exhibited works. This project will dramatically enhance public and University access to the holdings and significantly increase their utility for classroom teaching and research. The project will be jointly managed by Peter Hirtle (DAC Assistant Director) Carol DeNatale (Museum Registar). Joint selection and implementation of an image and object management system by major image repositories across the campus will also be conducted. Projected total costs are currently estimated at about $550,000.
This project includes major new challenges. Technical issues include the difficulties in effectively capturing digitally large artworks and three-dimensional objects and doing this in production environment. There are also special challenges in supporting effective display and use of these same items. It is hoped that some of the new technologies from IBM-Japan being tested in the Global Digital Museum may be employed in multi-directional viewing of sculptures and artifacts. Curators will have to make numerous selection decisions regarding items for inclusion without previous experience in the compilation or use of such a resource.
During the next two years, staff of the Interactive Multimedia Group and the Digital Access Coalition will assist the Cornell Plantations in the systematic employment of digital technology to integrate taxonomic and historical data with real-time field study. This project will expand public and academic access to this important botanical collection, enhancing its utility, linking it to related collections, and testing hand-held and wireless devices for field study and public touring. Preliminary planning has begun on a related pilot project to improve access to botanical specimens and historical data held by the L. H. Bailey Hortorium, linking this data to the developing Plantations database.
In the initial review of the Plantations database, I was struck by the importance of experienced designers in developing these Web collections. Questions of color, layout, and the selection and placement of navigational icons are critical to user appeal and effective use. While multimedia design is a rapidly expanding field, how many museums or libraries employ staff with this training and experience?
SagaNet is a joint project of the National and University Library of Iceland and Cornell University to establish and develop the founding component of an Icelandic National Digital Library. This cooperative effort will draw on holdings which in combination form the world's most extensive collection on Icelandic history, language, and literature. This project will create an electronic database and provide world-wide access to a remarkable research resource for the study of the Icelandic Sagas and epic poetry and for Scandinavian medieval studies.
The project will create high-quality digital images of the full texts of manuscript copies of Icelandic Sagas and rímur poetry (epic poetry on Saga motifs) and of printed editions and translations of the Sagas, as well as relevant critical studies published before 1901. Grayscale and color scanning will be employed in the conversion of vellum and paper manuscripts held by the National and University Library of Iceland and the Árni Magnússon Institute of Iceland, and high-resolution bitonal conversion from microfilm will be used for most of the printed volumes held by the Cornell University Library. In total, some 380,000 manuscript pages and 145,000 printed pages will be converted.
SagaNet will provide a unique model for the cooperative development of national digital libraries with international audiences, illustrating the potential that these technologies offer for the building of virtual collections accessible globally. The project will test the technical, economic, and qualitative viability of a model for international remote access to digital collections based on simultaneous, coordinated searching of existing on-line public access catalogs. This approach will utilize the capabilities of the International Standards Organization (ISO) Z39.50 standard in combination with World Wide Web technologies to provide integrated searching of distributed, and in this case, multi-lingual library catalogs and to furnish relevant full-text resources.
The utility of this database in supporting research, classroom, and public use of the full text of a definitive body of cultural literature will be systematically evaluated. Testing will incorporate both qualitative and quantitative evaluation. Quantitative studies will investigate both micro- and macro-economic issues, exploring the potential effects of digital library development on broader social, political, and economic conditions. Projected costs total $1.3 million, and if fully funded, this project will be a groundbreaking effort in the development of international research collections.
There are still some who see networked access as a threat replacing onsite research use to the detriment of our mission and rationale for existence. I am certain of the continued importance of our curatorial, preservation, and interpretation roles in maintaining our cultural and scientific heritage and feel that networked access will both advertise our holding and promote their value. However, while I am convinced of these points, it does not really matter in the same way anymore. We are working in a new environment in which our success is in part dependent on the proficiency with which we provide networked access and interpretation of our holding and on how effectively such access is in shapinga new profile for cultural and scientific repositories. A transformation is in process; the rules and rewards are changing; and we must imagine new models, devise new strategies, and employ new techniques.
Copyright Archives & Museum Informatics, 1997
Last modified: February 28, 1997
This file can be found below http://www.archimuse.com
Send questions and comments to firstname.lastname@example.org