Published: March 15, 2001.
Building a Web-based Collaborative Database--does it work?
Nuala A. Bennett, Trevor Jones Digital Imaging and Media Technology Initiative University of Illinois at Urbana-Champaign, USA
Designed to create a model collaborative environment, the Digital Cultural Heritage Community Project (DCHC) focuses on the digitization of materials from Central Illinois museums, archives and libraries for integration into elementary grades' social science curricula. Using teachers' curriculum units and the Illinois State Board of Education Learning Standards for Social Sciences as a framework, curators, librarians and archivists identify primary source materials from their collections for inclusion in a Web-based database to be used in third, fourth and fifth grade classrooms. The DCHC provides a means for museums, libraries and archives to identify common ground among their collections, and to provide schools with digital access to these materials. The Early American Museum (EAM) joined the DCHC out of a desire to digitize and increase electronic access to its collections. In this paper, we will discuss both the initial aims and goals of the DCHC from the standpoint of a large university library responsible for administering the project, as well as from the perspective of a smaller participant museum. We will discuss the artifact selection process for the database, and the implementation of a single database for a wide variety of collections, with particular reference to how the database design was viewed by the Early American Museum. Finally, we will explain how the project worked in reality.
Designed to create a model collaborative environment, the Digital Cultural Heritage Community Project (http://images.library.uiuc.edu/projects/dchc) focuses on the digitization of materials from Central Illinois museums, archives and libraries for integration into elementary grades' social science curricula. The ultimate goal of the project is to enable early and late elementary grade teachers in the locality to integrate digital cultural heritage materials into their curriculum, utilizing the Illinois State Board of Education Learning Standards. The project, which is administered by a unit of the University of Illinois Library at Urbana-Champaign, involves five other museum, archive and library partners and three elementary schools.
The Digital Imaging and Media Technology Initiative (http://images.library.uiuc.edu) administers the Digital Cultural Heritage Community (DCHC) project. The Digital Imaging and Media Technology Initiative (DIMTI) is a part of the University of Illinois Library at Urbana-Champaign. The unit was originally established in 1994 to perform research with collections and users to determine the best methods of providing digital access to the University Library's collections, The overall goals of the Digital Imaging and Media Technology Initiative program include: utilizing digital methods to preserve and to make accessible fragile and under-utilized visual resources; to promote the use of digital media throughout the campus and scholarly community; and to conduct research that advances the creation and use of these resources. These goals are met by establishing "best practices" for digitizing various classes of visual and textual materials, by developing multimedia databases that deliver visual resources and other media in innovative ways, and by conducting research on the ways in which visual information (photographs, drawings, illustrations, etc.) is used in the digital environment.
Other museum and library participants in the DCHC include the Illinois Heritage Association, the McLean County Museum of History, Lincoln Trail Libraries System, the Rare Book and Special Collections Library at the University of Illinois and the Early American Museum. The Early American Museum (http://node-02.advancenet.net/~early/), which is accredited by the American Association for Museums, is located in Mahomet, Illinois and is a facility of the Champaign County Forest Preserve District. It has an extensive collection interpreting 19th- and early 20th-century life in east-central Illinois. Two floors of exhibits present information on architecture, agriculture, trades and occupations, decorative arts, and childhood and domestic life in the past. The Discovery Room offers hands-on opportunities for children, and educational programs are offered for all ages throughout the year. The museum has an annual budget of under $250,000 and a full time staff of three. The museum's mission is to collect, preserve and interpret the history of East Central Illinois, specifically Champaign County, for the education and enjoyment of present and future generations.
In this paper, we will discuss both the initial aims and goals of the DCHC from the standpoint of a large university library responsible for administering the project, as well as from the perspective of one smaller participant museum, the Early American Museum (EAM). EAM joined the DCHC out of a desire to digitize and increase electronic access to its collections. The EAM curator was excited by the project's goal of putting "primary materials at teachers' fingertips through the Internet"  and hoped that it would provide an effective way for the museum to make its artifacts available in local classrooms. The experiences or comments of other curators or librarians participating in the DCHC might differ from those of the museum's curator, but this paper will only discuss the EAM experience.
Before the proposal for the project was finalized and submitted to the IMLS for funding, the director of each participating museum and library was asked to sign a letter of commitment agreeing that a staff person would work on the project for a minimum of two hours a week. Similarly, school principals also signed a letter showing that they understood the project would involve teacher time of two hours a week. In order to monitor each partner's contributions, time sheets were distributed and each partner submitted their hours to the Project Coordinator on a monthly basis. There was also an online version of the time sheet so that those who wished could submit their hours online.
It was appreciated by DIMTI that it was up to each partner to fulfill their obligation to the project, and that the work might require more than two hours a week. At the EAM, the curator sometimes spent much longer than two hours in one week doing work for the DCHC. However, while some weeks the curator would spend a lot more time than two hours working on the project, during other weeks he would put in considerably less time. At the end of the year, he calculated that he had spent approximately 100 hours over the course of a full year, which averages out to slightly less than two hours a week.
The EAM is open on a seasonal basis throughout the year. Because of this, the amount of time the curator spent working on the project did not necessarily detract from other projects at the museum. The biggest issue in terms of time was trying to create enough free time for extensive research that would benefit the DCHC. Although the process of digitizing information was not particularly time-consuming, the curator needed larger blocks of time to consider the interpretation of artifacts for the database. Experience now tells us that it would have been easier for the curator if artifacts chosen for the DCHC database could somehow also have been coordinated with exhibit development at the museum.
As at most small museums, artifact research at the Early American Museum is almost always undertaken only in conjunction with the exhibit development process. The museum usually completes one major exhibit project each year. If, for example, the museum were working on an exhibit about growing up in the early 1900s, it would have been easiest for the museum to digitize and interpret artifacts related to this theme. The problem of course, is that the museums and the teachers, as well as the libraries, are rarely flexible enough to change their plans for a specific project like the DCHC. Collaborative projects, however, that served both the teacher's educational objectives and the museum's exhibition objectives could potentially serve both parties well. This type of collaboration was beyond the scope of our project, but it raises intriguing possibilities for the future.
Choosing Materials for Digitization
The DCHC was set up to provide a means for museums, libraries and archives to identify common ground among their collections, and to provide schools with digital access to these materials. One of the first aspects of the project was to choose the collections that would be digitized for the online database. The teachers participating in the DCHC project were teaching third, fourth and fifth grade students. In order to give a sense of what kinds of artifacts and items the teachers would like in the database, they first shared with the museum curators, librarians and archivists, their curriculum units for social sciences (http://images.library.uiuc.edu/projects/dchc/resources.htm).
Illinois Learning Standards
In Illinois schools, the teachers must adhere to learning standards, as recommended by the Illinois State Board of Education. Adopted in 1997, the Illinois Learning Standards for Social Science (http://www.isbe.state.il.us/ils/socscience/social.html) outline expectations for student learning in the classroom from early elementary through high school education. The standards are a means to assess student achievement and "define what Illinois citizens believe all students should know and be able to do as a result of their public schooling" . They also help to provide fair and equitable education for all students throughout the state.
The teachers also included the Illinois State Board of Education Learning Standards for Social Sciences appropriate to each curriculum unit. The third grade curriculum units were more appropriate to the early elementary level of Learning Standards, while the fourth and fifth grade units corresponded to the late elementary Learning Standards. Curriculum units covered a broad spectrum within the social sciences. The Third Grade units ranged from "How we learn about communities", "Communities and their geography" to "Celebrations and Festivals", while the Fourth Grade units included "French in Illinois", "Prehistoric Indians of Illinois" and "Government in Illinois." Fifth Grade units covered topics from "The English Colonies", "U.S. Constitution and Government", "Westward Expansion" to "Revolutionary War."
Using the Illinois State Board of Education Learning Standards for Social Sciences as a framework, and the teachers' curriculum units as a guiding source, the curators, librarians and archivists then identified primary source materials from their collections for inclusion in a Web-based database to be used in the third, fourth and fifth grade classrooms. They were to choose materials that would be particularly appropriate to the curriculum units submitted by the teachers. At DIMTI, we had anticipated that there would be many artifacts in each of the participating museums and libraries that could be digitized for use in the classroom by the teachers.
Deciding what to digitize for the project proved to be a major challenge for the EAM. Within their curriculum units, the teachers had some specific requests, but they were unsure of the type of things museums had in their collections and how those artifacts could help with their classroom teaching. The curriculum units each listed a theme, the time spent on that theme, and the major historical events and figures addressed. While the units provided the museum with some new ideas, they also presented new problems. The units primarily focused on state and national history, and only infrequently made use of local examples.
Using the state's Learning Standards for Social Studies alone to determine what artifacts to digitize would also prove very difficult to implement. By using the standards, museums and libraries should theoretically have been able to digitize artifacts that would mesh with social studies curricula not only in the participant classrooms but also throughout the state. A more challenging obstacle for determining what to digitize turned out to be the vagueness of the standards themselves. Designed to provide a broad framework for teaching, the standards include very few specific recommendations, making it arduous to determine which specific artifacts in the museum's collection could serve each particular learning goal. At the same time, the curator could also think of several hundred other artifacts that might serve equally well. The scope of the standards was so broad that it was necessary to find a way to narrow down the list of potential artifacts.
The vast scope of the curriculum units again created problems for the EAM. The museum focuses on the history of East Central Illinois, and consciously works to interpret the lives of ordinary people who do not often show up in history textbooks. As a result, the collection includes few objects of state or national significance, and has a limited chronological scope. The collection includes some artifacts pertaining to Native Americans and the expansion of the west, but the majority of artifacts were made after Champaign County, Illinois (the county where the EAM is situated) was founded in 1833. The museum collections could easily support the "Westward Expansion" unit, a small section of the "French in Illinois" and "Revolutionary War" units, but very little of the others.
The museum had problems not only finding artifacts for the teachers, but also in answering other requests which they posed. To the EAM curator, the teachers often seemed to be seeking not simply images of artifacts, but rather entire educational components which could be used to expand their curriculum. Their requests for "a pictorial timeline of different transportation methods" for a unit on "American Communities in History", or "theories about the decline of the Hopewells and the Mississippians" for a Native American unit made it clear that the teachers were looking to the museums and libraries for answers that required substantial historical interpretation.
Through subsequent online discussions with all project participants, the museum curators, archivists and librarians suggested artifacts from their collections that might be digitized to match the teachers' curriculum units. They also suggested resources that the teachers might not have considered using, but which the curators and librarians felt matched the Illinois State Learning Standards corresponding to the teachers' units. Artifacts were eventually chosen for digitization through extensive online discussions with the teachers. In the end a substantial number of artifacts were chosen not because of the teachers originally requested them, but because the museum curators and librarians suggested items to the teachers who in turn agreed that they would be useful.
Digitization of Materials
DIMTI organized a scanning workshop to ensure that all museum curators and librarians would be able to produce quality and consistent images for use in the final online database. The focus of the workshop was the capture of print or graphical materials and creation of digital files, as opposed to other aspects of scanning such as file storage, object description or intellectual property issues. Participants were introduced to scanner operation, the use of scanning utilities, image manipulation software, resolution, color-depth, and quality indices as well as calibration of scanners and monitors for optimal digitization. They also learned about the general guidelines for digitization as presented by the Library of Congress  and also, using the Cornell University digitization formulae . The Library of Congress guidelines cover formats used for archiving and for Web access to their American Memory collections (http://memory.loc.gov/). The Library of Congress guidelines are an "attempt to balance reproduction quality, convenience of access, likely longevity of the format itself, production cost, and a preference for true standard formats or for industry formats that have been widely adopted".
Guidelines have been set up at the Library of Congress for three categories of image -- thumbnail, access and archival. For example, the Library of Congress guidelines recommend that thumbnail images of photographs be 72 dpi and 150-200 pixels across the long dimension. The access image is a higher quality image that has been compressed for speed of access and has a lower spatial resolution than the archival image. In turn, the archival image is an uncompressed image, which would be useful for users for reproduction efforts or held for future reprocessing as compression or other image-processing standards might change. In the case of a photograph, the image should be 3,000 to 5,000 pixels across the long dimension. The Library of Congress recommends guidelines for different types of artifacts. For the DCHC, we concentrated on those guidelines for scanning text documents, photographs, maps, and graphic materials such as artwork.
Although all of the project participants agreed to adhere to the standards, they had little experience of digitizing for archival purposes or for specific online databases, and it took time for some participants to become familiar with the requirements for the DCHC database. Creating archival images at local museums and libraries would take up a lot of computer storage, and take up costly space to download the images directly into the database over a modem connection. Therefore, participants were also supplied with CD writers in order to able to burn their images onto CD. The complete CDs were subsequently mailed to the Digital Imaging and Media Technology Initiative in order to download the images into the database. It is worth pointing out that for a bigger project, this would create a lot of CDs for storage, so other storage facilities might be investigated.
The participants simultaneously determined the steps necessary to make the final digital content available to teachers and students, and to create the digital repository that would be accessible through the project Web site. In order to create an integrated system usable by each participant institution, the use of metadata for the image database was of major concern. The next workshop held in conjunction with the DCHC was a daylong session to address metadata issues. Participants came to terms with metadata topics and then discussed their institutional concerns about metadata. For example, the museum participants were all very concerned about interpretive comments about artifacts, while library participants were less concerned about interpretation. We believe all DCHC project participants would agree with Abby Smith from the Council of Library and Information Resources, when she said, "Museums ... have a mandate to interpret. Librarians ... traditionally place a high value on making information accessible without mediation" .
DIMTI had to develop a set of guidelines and standards for the museums and libraries to enable the creation of a consistent database that would easily map between different controlled vocabularies and organizational data. It was decided to base the fields in the database on the Dublin Core (DC) fields, but due to the diversity of the partners and their collections, some adjustment of the original fifteen DC fields was necessary. Figure 1 shows a sample record from the database with each of the fifteen fields shown in bold. Almost none of the participants had worked with DC before their involvement in this project, so it was imperative to set very specific guidelines to enable the entering data into the database with complete consistency.
Dublin Core is a set of fifteen elements, or fields, that are intended for facilitation of electronic resources. This element set has several characteristics that made it ideal for the DCHC. In particular, its relative simplicity enabled us to ensure that all project participants could quickly grasp the relevant issues. Its semantic interoperability also made it a prime candidate for project users at museums and libraries, with very diverse collections. Finally, the DC extensibility also allowed for additional metadata elements, which could encompass the variety of the participating institutions. (There is a more detailed discussion of the database fields in ).
Interpretation of Artifacts
While the DCHC database was designed to digitize and make accessible items from the partner institution's various collections, it was not specifically intended as an aid to interpreting the meaning of these artifacts for teachers. Digital and cataloging expert Jim Blackaby wrote in his letter of support  for the DCHC that the project "focuses on making primary materials available in a context that is enabling. It is not the delivery of interpreted materials, such as an exhibit, which is momentarily satisfying and may address immediate needs, but in the long run may lose some of its efficacy. Where libraries and archives have focused on building a solid core of information systems that provide access to primary materials, museums have not. In this respect, those institutions and their methods bring museums along".
The decision originally not to build an interpretive component into the original DCHC design proved to be very frustrating for the EAM and reflects philosophical differences between museums (especially history museums) and libraries. Historians believe that historical events must be placed in a broader context in order to have meaning, while libraries often feel that their primary role is to provide objective access to information. Museum professionals in general are keenly aware that it is almost impossible to provide objective information in the museum context. The process of choosing which objects to exhibit, how they are labeled, and the decision whether to mount them at front or the back of a case all work to tell visitors what we think is important for them to know. In the words of M. Jane Young, simply placing an object on a shelf "changes the shape of the object, divorcing it from contextual relationships, isolating it from experience, and thus largely predetermining the visitor's perception of it" .
While it is also true that libraries interpret data, their interpretive functions are less pervasive than in museums. Libraries interpret by choosing which books to purchase or not purchase and by recommending titles to their patrons. Regardless of whether interpretation is typically a library function, it became one for all of the DCHC partners during the course of the project. All of the partner institutions performed a clear interpretive task by choosing which artifacts, photographs and maps to digitize and place on the DCHC database.
Interpretation is much more powerful and useful, however, when interpretation of artifacts is acknowledged as important, and a conscious effort is made to do it well. Good interpretation does not mean telling visitors or patrons what to think, but rather provides a basis for them to ask questions and learn answers for themselves. Museums can play an important role in this process by providing access to artifacts and contextual information about them. In general people "respond more directly and immediately to objects than to verbalized concepts. Artifacts do seem to inspire, excite, motivate and communicate" . This concept of interpretation of artifacts was very new to almost all the librarians participating in the DCHC.
At the EAM, the goal is to help teachers use artifacts to inspire and motivate their students to get excited about history. An important part of the museum's educational role is to provide teachers with the type of specialized historical information they might not have the time or expertise to gather. The museum curator sees artifacts as useful in that they help to communicate ideas to the public. The power of artifacts stems from the stories they can tell about the lives of people in the past, and our lives in the present. Originally the DCHC database made no provision for communicating these stories to the teachers who would use the artifacts in the classroom. All of the partner teachers wanted to use artifacts to teach their students, but most knew little about 19th century material culture or how to effectively incorporate it into their lesson plans.
An artifact from the DCHC database, as seen in Figure 2, helps illustrate the importance of contextual information. One artifact the Early American Museum digitized was a bed key. The description in the DCHC database states that it is made of wood, is 12.25" long, consists of a Y-shaped shank with a round cross bar pegged near the top, and was made between 1775 and 1830. This information is descriptive, but it does not help elementary school students learn about the past. Adding interpretive information might make a digital image of this artifact more useful: "In the 19th century many bed frames used ropes to hold up mattresses which were often filled with corn husks. It was hard to sleep well on a saggy mattress, so periodically people had to tighten the ropes in order to sleep well -- this is where the phrase "sleep tight" comes from." This snippet of interpretive information connects the past to the present and helps the artifact come alive. By using the artifact as an introduction, a teacher could use it as a springboard to discuss the cycle of a pioneer's day, sleeping habits, notions of privacy, the importance of corn in pioneer life, or any number of other things.
Although the original DCHC concept contained no provisions for interpretation, the project's database, based on Dublin Core, was modified to include a separate interpretive field. The new field provided a method to add the interpretive information, which the EAM feels is so crucial. The addition of Interpretation was considered a major improvement to the database, but additional work could still be done. Even with the interpretive field in place the majority of the artifacts in the database are still expected to "speak" for themselves, and most of the information entered into the interpretation field remains superficial. Ideally, the museum curator would like interpretation to be a central part of the DCHC database. Each digitized image could be tied to an appended list of potential sources, links to other sites, comparisons with similar objects, and tie-ins to larger historical themes and arguments.
Unfortunately, the problem with good interpretation is that it takes a lot of time. Writing even a short descriptive paragraph with minimal interpretive information for each digitized item is a painstaking process, and placing each item into a broader historical context would be an even greater challenge. The amount of time necessary to do this could run into many more hours than the two hours each week originally agreed to by all participants. Nevertheless, the EAM curator thinks it is worth doing. Putting images in a database facilitates access to museum and library collections, which may be useful to researchers, but it can fail to make history come alive for elementary school students. If projects like the DCHC are truly more interested in assisting educators do a better job teaching American history, rather than simply providing access to information, we need to spend more time making what we digitize useful for teachers and students and less time worrying about getting it on the Web. In fact, every participant of the DCHC feels so strongly about the importance of using the database for education that we are planning for 2001-2002 to work on a new project, "Teaching with Digital Content--Describing, Finding and Using Digital Cultural Heritage Materials" (http://images.library.uiuc.edu/projects/tdc), where we hope such issues can be addressed.
Agreements between Partners
Intellectual property issues involved in the digitization of images include the right to distribute materials, permissions to host materials, institutional ownership, content that could be offensive to some users, and conditions of use. Because the DCHC partners were concerned with these matters, the partners worked to develop a set of terms and conditions into a Collaborative Agreement for digital access to their collections.
The agreement ensures that there are no misunderstandings among partners either during the course of the project or after it is officially completed. Project participants agreed to provide selected images and accompanying information in digitized form. The agreement outlines procedures for making the digitized images, descriptive text and other media available to participating institutions in agreed upon formats, and confirms how works are selected for inclusion in the project. The agreement also covers copyright issues, such as whose responsibility it is to obtain permissions to digitize objects, and it confirms that images and their accompanying metadata remain the intellectual property of the partner.
As well as the collaborative agreement for each partner institution, a "Conditions of Use" (http://images.library.uiuc.edu/projects/dchc/conditions.htm) statement for the project Web site was also created. In this case, the statement was created for the protection of the project participants and the database users.
The EAM supports the commitment to maintain the database for at least two years beyond its official end date. The EAM was not particularly concerned with possible copyright violations, as it does not generate revenue from its collection the way that a larger museum might. A larger institution might be concerned with possible revenue losses if the database was maintained for a longer term. The EAM is, however, concerned about the interpretation of artifacts and the use of the interpretation as it sits on the DCHC web site. This will have to be monitored as closely as possible over the next few years.
From the perspective of the university library, the project has been a success. The DCHC was set up as a "Model Program of Cooperation" and in that perspective it was an excellent example of cooperation among several diverse institutions. Work was carried out collaboratively in various forms. There were several workshops and meetings with all project participants, as well as many on-site visits from project staff. Communication was also carried out by telephone, email, initially using a Listserv and subsequently switching to an online web-based discussion board (WebboardTM). Everything that was initially planned for the project was carried out. The project began with a timeline of activities set out for the course of the project. While the activities on the timeline may have moved during the project lifespan, the actual activities were all completed, some more or less than initially planned. Some parts of the project, such as selection of artifacts and entering them into the database took considerably longer than anticipated. Nevertheless, the database took shape and became available to the teachers for use in their classrooms.
The true test of the usefulness of the database itself is how it is used by teachers and students. The most exciting part about the DCHC is that it can make it easier for teachers and students to view a diversity of collections in their own classroom. We are currently carrying out more research into how the teachers actually use the database images in their curriculum. As part of this research, future extensive outcome-based evaluation studies will help us ascertain how much use the database is actually receiving from our local users as well as from users worldwide. Further work will be done to develop innovative ways in which teachers may further develop their curriculum units using the images in the database. In order to continue to push the boundaries of the project we have also have enlisted the support of an Advisory Group of eight experts from various fields who provide suggestions on how to improve the DCHC in the future.
This work is sponsored by the Institute of Museum and Library Services under its National Leadership Grant Program.
 Grant application to the Institute of Museum and Library Services for the Digital Cultural Heritage Community Project, 1998.
 Illinois State Board of Education, Illinois Learning Standards, http://www.isbe.state.il.us/ils/, last viewed Jan. 10, 2001.
 Carl Fleischhauer (Last updated July 13, 1998), Digital Formats for Content Reproductions, Library of Congress. Last viewed December 19, 2000. http://lcweb2.loc.gov/ammem/formats.html/
 Anne R. Kenney, Stephen Chapman (June 1996), Digital Imaging for Libraries and Archives. Ithaca, NY: Cornell University.
 Smith, Abby (January 2000), Putting Culture Online, in Collections, Contents, and the Web. Council on Library and Information Resources.
 Bennett, N.A., B. Sandore, A.M. Grunden, P. Miller (2000), Integration of Primary Resource Materials into Elementary School Curricula, in Proc. MW2000, pp. 31-38.
 Jim Blackaby (April 17, 1998), "Support for the IMLS Project", Digital Cultural Heritage Community Grant Application to the Institute of Museum and Library Services.
 M. Jane Young (1987), "The Value in Things: Folklore and the Anthropological Museum Exhibit". In Folklife and Museums (p. 103.), Patricia Hall and Charlie Seeman, eds. Nashville: AASLH.
 Michael J. Ettema (1987), "History Museums and the Culture of Materialism". In Past Meets Present: Essays about Historic Interpretation and Public Audiences (p. 77), Jo Blatti, ed. Washington D.C.: Smithsonian Institution Press.