The community digital library has emerged as a recent evolution in the development of digital libraries. A community digital library is distinct through having a community of potential users define and guide the development of the library. In this article, we present how the geoscience community has worked to develop The Digital Library for Earth System Education (DLESE) in the light of recent work in the area of meta-design, the design of the design process. The development of DLESE is described utilizing a conceptual framework developed from the analysis of a variety of open source projects.
Community libraries, more than any other civic institution, are the hallmark of an informed society. Libraries carry powerful symbolic meaning and inference. A library can serve as the communal repository of knowledge, the town meeting place, a touchstone for intellectual discussion and community activism. A community library is a free and accessible place, hospitable to both the learned scholar and the struggling student. A library has the potential to transform the way a community thinks about itself; it provides the conditions and possibilities for intellectual growth, social mobility, and community development and cohesion.
Community libraries are evolving with the information age. The past decade has witnessed the increasing ubiquity of the World Wide Web in homes and schools, the emergence of new kinds of "electronic communities" (Preece 2002), and the widespread creation and distribution of digital educational materials. Digital libraries have emerged as a means for disciplinary communities to share, organize, and assess their intellectual holdings. To date, there are many digital library efforts underway aimed at improving K-12 and undergraduate science education (e.g., NEEDS, iLumina, TeacherLib, DLESE). One prominent example is the NSDL (National Science, Technology, Engineering, and Mathematics Digital Library), a program initiated by the National Science Foundation (NSF) to integrate multiple digital libraries and to serve the needs of the community of science educators and learners at the broadest level (Fulker and Janée 2002).
Over the past three years, the Earth system science community has come together to articulate a vision for, and begin construction of, The Digital Library for Earth System Education (DLESE). DLESE is being designed to support the specific educational needs of the geoscience community within this larger national library network. In the tradition of community libraries, the mission of DLESE is to fundamentally change the way students learn, instructors teach, and researchers interact, by providing new ways of sharing information, tools, and services.
DLESE is a grassroots, community-led project that, among a range of services, provides searchable access to high-quality, online educational resources for K-12, undergraduate, and informal Earth system science education (Marlino, Sumner et al. 2001). These resources include objects such as maps, lesson plans, lab exercises, data sets, virtual field trips, and interactive demonstrations. The holdings of DLESE are created by a wide variety of individual faculty members, agencies, and institutions. These resources are held (stored) on local servers and are accessed through the library via a database of searchable metadata records that describe them.
To date, significant progress has been made on many aspects of the library: the community has been organized, a governance structure has been established, a strategic plan has been developed, a useful collection is available, and a working version of the library is now in use (Figure 1). This operational library was released in August 2001, and contains approximately 2000 carefully selected educational resources. Version 1.0 provides educational discovery features that enable users to search by grade level, educational resource type, and keyword. This version of the library also contains a web-based Resource Cataloger enabling community members to contribute resources to the library. Additional community oriented services are provided such as discussion forums for working groups, and posting tools for geoscience education opportunities and announcements. Future versions of DLESE will offer peer-reviewed teaching and learning resources, interfaces and tools to allow exploration of Earth data, and services to help users effectively create and use educational resources.
Figure 1. Version 1.0 of DLESE (www.dlese.org) was released in August 2001. It enables educators and students to search and browse for educational resources by grade level, keyword, and educational resource type. Every month an educational resource created by a community member is featured.
Designing the Design Process
The uniqueness of community libraries lies not in whom they are being built for, but in how they are designed, constructed, and managed. A library such as DLESE cannot thrive without the distributed effort and energies of a broadly engaged community. Our experiences to date demonstrate the benefits of adopting a design process where community-building and technical development are tightly integrated: through workshops, working groups, and funded projects, community members are actively participating in the development of library policies, technologies, collections, and peer reviewing services.
DLESE's unique approach to distributed library design and construction is informed by the emerging theoretical perspective of meta-design (Fischer 2001; Fischer 2001). Meta-design is a useful perspective for analyzing projects where �designing the design process' is a first-class activity, i.e., creating the technical and social conditions for broad participation in design activities is as important as creating the artifact itself. Meta-design theory draws on existing design traditions such as user-centered design and participatory design.
Methodologies such as user-centered design (Norman 1986) and task-centered design (Gould, Boies et al. 1991; Lewis and Rieman 1993) emphasize getting systematic user feedback on the evolving design throughout the design process; however, it is largely designers who generate possible solutions, users are mainly in a reactive role. Participatory Design (PD) methods (Greenbaum and Kyung 1991; Schuler and Namioka 1993) seek to involve users more deeply in the process as co-designers; i.e., users also propose and generate design alternatives. An important ideal of PD is industrial democracy (Ehn 1989), which implies that users should be politically empowered with a broad design remit that includes: co-designing the technology, co-determining the nature of the design process itself (schedules, processes, outcomes), and co-designing the social conditions of technology use (i.e., work practices, organizational structures and responsibilities).
Meta-design is largely an empirically inspired theory, derived from examining success models of open source projects, where the design and development of complex software systems is carried out by highly distributed, potentially large groups of people communicating and coordinating activities via the Internet (Raymond 2001). While meta-design shares the industrial democracy ideals of participatory design, the social and technical conditions under which design is performed, as well as the anticipated outcomes, are quite different:
A detailed analytic framework for meta-design based on an analysis of open source projects was recently developed by Scharff (Scharff 2002). The framework illustrates the collaborative process toward developing an object, in Scharff's case, a piece of software where the participants are a group of software developers (Figure 2). A core principle of the framework is the intertwining of social processes and technical artifacts. The participants use various computer-mediated communication technologies for social interaction (email, news, web forums) and software development tools that support collaboration and coordination (e.g., CVS, or Concurrent Versions System, a tool used to manage changes to software code). The participants use these technologies to support their collaborative process to contribute to the development of the final object, or software. We should note this framework also captures the iterative nature of development: participants can contribute to the object creation process after using interim versions, and where the state of the design is accessible usually by means of a public artifact or standard distribution.
Collaborative Library Development in DLESE
In the previous section, we describe how meta-design theory has been derived from the examination of open source projects. Some common traits across the projects are that participants are largely volunteers, distinctions between developer and user roles are blurred, participants are numerous and highly distributed, and various sub-projects can emerge and proceed in parallel, coordinated by a small team or an individual. These traits are also evident in DLESE with its highly distributed community, community members being users and developers, and various DLESE sub-projects being developed in the community with policy oversight and coordination through a community-based governance mechanism. This governance mechanism formally adopted open source principles for DLESE in December 2000. Because of these similarities, among others, we believe that Scharff's framework is highly relevant for analyzing the distributed library construction process.
In the case of DLESE, the participants are a broad collection of engaged individuals from the geoscience community (researchers, educators, and students), K-12 educators, library science specialists, digital library and information sciences specialists. The participants are distributed across the USA. These participants can be arranged into three primary groupings: governance, the DLESE Program Center (DPC) and the community. Governance includes the steering committee responsible for overall policy guidance for the DLESE project, and four standing committees aligned to core facets of the library (services, users, collections and technology). The DPC is responsible for developing and operating the core infrastructure of the library, and provides a coordinating role for the DLESE project. This coordinating role is similar to the coordinating roles of teams in other large open source projects (e.g., the Apache coordinating team on the Apache project (Fielding 1999), or Squeak Central (Kaehler 2002; Scharff 2002)). The community includes the individuals and institutions that have an interest in seeing DLESE develop, and they are involved through individual action and through structured arrangements such as working groups (reporting to standing committees) discussion groups, and events such as the DLESE Annual Meeting. As noted earlier, individuals contribute as project collaborators or leaders based on their interests and expertise.
The collaborative process in DLESE has allowed the community to interact and participate in the design and development of the library. The framework highlights two facilitation mechanisms to support the collaborative process: a social facilitation of participants; and a technical facilitation of communication infrastructure and technical artifacts. The overall process toward developing the library (the object) has involved both these paths.
An important aspect of the DLESE process has been the social facilitation of participants through face-to-face meetings and workshops, as well as through the technical facilitation using mechanisms such as computer mediated interaction, e.g., email, web-based forums, and document sharing. The social facilitation has been very important to the DLESE development in that it helps community members from different backgrounds understand and share a common vision of what DLESE should be. In addition, explicit use of user-centered and participatory design methodologies to support the collaboration process has proved to be very useful. This has included the development of use cases derived from envisioned use scenarios written by educators and from interviews with prospective users (Davis and Dawe 2001; Sumner and Dawe 2001). As a check on library development progress, formative usability testing is done to ensure that the library development is meeting the goals (as expressed through the use cases), and to provide further contributions to the development. The development of use cases and the testing of the library against them is a level of formalization on the contribution and feedback mechanisms beyond those found in the open source projects analyzed by Scharff.
DLESE has taken advantage of various computer mediated communication mechanisms (email, web forums, document versioning, development and discussion of use cases) and human mediated mechanisms (face-to-face meetings, telephone conferencing and workshops). For any given component project within DLESE, a number of these collaborative technology mechanisms have been used, often starting with work group meetings followed by extensive online discussions around shared documents. We have found that support for document versioning is crucial and needs to be available and accessible to all participants. We have found that the versioning systems commonly used in open source projects (e.g., CVS) do not lend themselves to use by our broader, non-software developer community.
The development of DLESE is an ongoing project, but services are already being provided to the geoscience community. Services include the core technical mechanisms of collection building and discovery, plus the community building support such as online groups and human interaction through meetings such as the Annual Meeting. There are projects now under development in the wider DLESE community that are building on this core, e.g., the Community Review System (CRS) under development at Columbia (Kastens 2001) and the Community Issues and Groups area at Carleton College (Manduca 2002). These projects utilize underlying technical structures developed by the DPC to explicitly support distributed library building.
Each of these developments has evolved from the early stages of the ongoing DLESE collaborative process. The leaders of these developments came forward from the larger DLESE community to lead the development of an identified library component and to coordinate their work with other projects. The subsequent development of the components have themselves elicited a collaborative process in their development through working with specific working groups and committees (e.g., the CRS with the collections committee, the metadata working group and the DPC), so the conceptual framework has applicability at different granularities of our library development process.
DLESE is a partnership between the National Science Foundation (NSF), the DLESE community, the Steering Committee, and the DLESE Program Center. The success so far in developing DLESE has been based on the effective collaboration of the DLESE participants to date. The framework developed by Scharff has given us a tool on which to build the description of DLESE development and to illustrate how the collaborative process has been important to DLESE development. While Scharff's work looked at distributed software developers on open source software development projects, we have found that the framework is useful in the context of a broader participant pool such as that found in DLESE. One major factor that other digital library developers heading down the community library path should consider is the large amount of work required to support the social facilitation of a diverse set of participants beyond core digital library technologists to allow all participants to have a shared vision, and allow them to participate fully in the design and development. In DLESE, the structure and use of meetings, committees, governance, and DPC has provided a framework (the meta-design) to support the distributed library building process.
This article has benefited from discussions with Eric Scharff, Gerhard Fischer and other colleagues at the Center for LifeLong Learning and Design. The research described here was partially funded under Cooperative Agreement #ATM-9732665 between the National Science Foundation (NSF) and the University Corporation for Atmospheric Research (UCAR), and Grants #9978338 and 0085600. The views expressed herein are those of the author(s) and do not necessarily reflect the views of NSF, its sub-agencies, or UCAR.
Davis, L. and M. Dawe (2001). Collaborative Design with Use Case Scenarios. ACM/IEEE Joint Conference on Digital Libraries (JCDL '01), Roanoke, VA (June 24-28), pp: 146-147.
Ehn, P. (1989). Work-Oriented Design of Computer Artifacts. Stockholm, arbetslivscentrum, 496 pages.
Fielding, R. (1999). "Shared leadership in the Apache project." Communications of the ACM 42(4): 42-43.
Fischer, G. (2001). Communities of Interest: Learning through the Interaction of Multiple Knowledge Systems. 24th IRIS Conference, Bergen, Norway (August), pp: 1-14.
Fischer, G. (2001). External and Sharable Artifacts as Sources for Social Creativity in Communities of Interest. Fifth International Roundtable Conference on Computational and Cognitive Models of Creative Design, Heron Island, Australia (December 9-13), In Press.
Fulker, D. and G. Janée (2002). Components of an NSDL Architecture: Technical Scope and Functional Model. To appear in: Proceedings of the ACM/IEEE Joint Conference on Digital Libraries (JCDL '02), Portland, Oregon (July 14-18), In Press.
Gould, J. D., S. J. Boies, et al. (1991). "Making Usable, Useful, Productivity-Enhancing Computer Applications." Communications of the ACM 34(1): 74-85.
Greenbaum, J. and M. Kyung (1991). Design at Work: Cooperative Design of Computer Systems. Hillsdale, NJ, Lawrence Erlbaum Associates, 294 pages.
Kaehler, T. (2002). Squeak, [http://www.squeak.org]. Last access: April 29, 2002.
Kastens, K. A. (2001). "How to identify the "best" resources for the reviewed collection of the Digital Library for Earth System Education in Column "Another Node on the interNet"." Computers & Geosciences 27(3): 375-378.
Lewis, C. H. and J. Rieman (1993). Task-centered User Interface Design: A Practical Guide, [http://home.att.net/~jrieman/jrtcdbk.html].
Manduca, C. (2002). DLESE Community Issues and Groups, Carleton College, [http://www.dlesecommunity.carleton.edu]. Last access: April 29, 2002.
Marlino, M., T. R. Sumner, et al. (2001). "The Digital Library for Earth System Education: Building Community, Building the Library." Communications of the ACM Special Issue on Digital Libraries (May): 80-81.
Nakakoji, K., Y. Yamamoto, et al. (2002). Evolution Patterns of Open-Source Software Systems and Communities. International Workshop on Principles of Software Evolution (IWPSE 2002), Orlando, Florida (May 19-20), In Press.
Norman, D. (1986). User Centered System Design, New Perspectives on Human-Computer Interaction. Hillsdale, NJ, Lawrence Erlbaum Associates, 31-62 pages.
Preece, J. (2002). "Special Issue: Supporting community and building social capital." Communications of the ACM 45(4): 37-39.
Raymond, E. (2001). The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary. Sebastopol, CA, O'Reilly & Associates, 255 pages.
Scharff, E. (2002). Open Source Software: A Conceptual Framework for Collaborative Artifact and Knowledge Construction. Ph.D. Thesis, Department of Computer Science, University of Colorado, Boulder: 181.
Schuler, D. and A. Namioka (1993). Participatory Design: Principles and Practices. Hillsdale, NJ, Lawrence Erlbaum Associates, 319 pages.
Sumner, T. R. and M. Dawe (2001). Looking at Digital Library Usability from a Reuse Perspective. ACM/IEEE Joint Conference on Digital Libraries (JCDL '01), Roanoke, Virginia (June 24 -28), pp: 416-425.
(On May 17, 2002, a link in the caption of Figure 1 to the article was corrected.)
Copyright © Michael Wright, Tamara Sumner and Mary Marlino