.
.	CNI FALL 1998 TASK FORCE MEETING HANDOUT	.
.

MoA II  Whitepaper Executive Summary Version 2.1  (October 15, 1998)

( Note: The complete whitepaper can be found at http://sunsite.berkeley.edu/moa2/ )

            The Making of America (MoA II) Testbed Project is a Digital Library Federation (DLF) coordinated, multi-phase endeavor that proposes to investigate important issues in the creation of an integrated, but distributed, digital library of archival materials (i.e., digitized surrogates of primary source materials found in archives and special collections). This paper is a milestone in the MoA II planning phase and identifies a starting point for the testbed that will be created in the production phase of this project, with funding from the National Endowment for Humanities. An overview of this paper's goals and the MoA II project is contained in this executive summary. Detailed project background information follows in the next section of this paper.

            The library community has a distinguished history of developing standards to enhance the discovery and sharing of print materials (e.g., MARC, Z39.50, ISO ILL protocols, etc.). This leadership role continues today through library participation in creating new best practices and standards that address digital collections and content issues (e.g., EAD, TEI, preservation imaging, etc.). In addition, libraries have worked actively within the broader Internet community to adopt other standards that are used to store and access digital library materials (e.g., TIFF, HTTP, URNs, etc.). Perhaps the most important goal of this paper is to open a new dialogue in the ongoing conversation about digital library standards, specifically, to discuss the need for any new best practices and standards that are required if the digital library is to meet traditional collection, preservation, and access objectives.

            The discussion this paper hopes to stimulate builds on work completed to date and asks the question, "How can we create digital library services that interoperate in an integrated manner across multiple, distributed repositories?" Clearly, the standards and best practices mentioned above play an important role in answering this question. However, this paper and the MoA II Testbed Project in general focus on a new area of discussion that goes beyond the discovery of a digital object, and focuses on how it is handled once it is found. That is, the paper and testbed focus on the need to develop standards for creating and encoding digital representations of archival objects (e.g., a digitized photograph, a digital representation of a book or diary, etc.). If tools are to be developed that can work with digitized archival objects across distributed repositories, these objects will require some form of standardization.

            This paper aims to begin the discussion of digital object definitions by developing and examining metadata standards for digital representations of a variety of archival objects, whether they be in the form of text, digitized page images, photographs, etc. For our purposes there are three types of metadata: Descriptive, Structural,  and Administrative.  Descriptive metadata  is used to discover the object. The project testbed proposes to use existing descriptive metadata standards (such as MARC records and the Dublin Core), as well as existing descriptive/structural metadata (like the EAD) to help the user locate a particular digital object. The paper proposes defining new standards for the Structural  and Administrative  metadata that will be needed to view and manage digital objects. Structural metadata  defines the object's internal organization and is needed for display and navigation of that object. Administrative metadata  contains the management information needed to keep the object over time and identify artifacts that might have been introduced during its production and management (e.g., when was the object digitized, at what resolution, who can access it, etc.).

            At a higher level, this paper proposes a Digital Library Service Model in which services are based on tools that work with the digital objects from distributed repositories. This borrows from the popular object oriented design model. It defines a digital object as encapsulating content, metadata and methods. Methods  are program code segments that allow the object to perform services for tools, such as "get the next page of this digital diary." Unlike other models, methods are included as part of the object. This paper proceeds by identifying several archival digital object classes  that will be examined as part of the MoA II project, including photographs, photo-albums, diaries, journals, letterpress books, ledgers and correspondence. One of the first development efforts for the testbed will be to create the tools that display and navigate these MoA II objects, some of which have complex internal organization. Therefore, another goal of this paper is to identify the structural metadata elements that are needed to support display and navigation, to ensure they are included as part of the digital objects. In addition, this paper begins to examine the methods (program code) that could be included with each class of object.

            Because each partner library in the MoA II project will digitize images, the paper also investigates issues around best practices for digitization, in particular the capture of administrative metadata as part of this process.

            After this paper has been reviewed by the wider community, the MoA II participants plan to incorporate reader feedback into the development of digital object definitions for the classes of materials to be examined in the MoA II Testbed. These definitions will specify how to encode the content, metadata and methods as part of the object. An important goal of the project is to use the testbed to investigate the advantages and limitations of these definitions and help stimulate a broader discussion of standards for digital library objects and best practices for digitizing archival materials. This discussion must include the project participants, the DLF membership and the wider community. In addition, the project will contribute to the existing discussion in the DLF Architecture Committee on distributed system architectures for digital libraries. The MoA II testbed will give the library and archival community a tool that can be used to test, evaluate and refine digital library object definitions and digitization practices. It is expected that these discussions will move the archival community and the library community in general, closer to consensus on standards and best practices in these areas.

webmgr@cni.org