Since I spent a good part of the week thinking about and writing the following proposal for the 2008 Open Repositories Conference, I thought I would double-dip once more when it came to posting to the blog.
Developing A Fedora-based, Enterprise Class, Digital Asset Management System For Cultural Institutions.
In collaboration with several departments throughout the university, Cornell University Library is beginning to develop a digital asset management system based on the open source Fedora repository that is specifically designed to meet the needs of cultural institutions. The project itself is described along with an analysis of the organizational challenges it seeks to address.
In cultural institutions, tremendous volumes of data–specifically, digital assets in the form of text, image, audio and video files–are created by faculty and staff that are inaccessible because that data is not organized and stored so it can be accessed and repurposed by others in the organization. Often this data is simply lost as the members of these institutions leave. This challenge not only applies to data in physical form that could be digitized, but includes the explosive growth of born digital content. To meet this challenge, cultural institutions can benefit from the use of a centralized, enterprise class digital asset management (DAM) system.
The DAM will provide the first opportunity to centrally collect, organize and store the organization’s digital assets, further facilitating quality control of the digital masters and more formal metadata creation. The DAM will lower technical and cultural barriers to centrally managing data produced by faculty and staff, giving them greater control over their data while increasing the potential value of that data to the institution. The DAM will then enable the institution to leverage these assets by providing opportunities to select assets for such purposes as research, publication and teaching as well as inclusion in institutional repositories and digital library collections.
Enterprise class DAM systems are designed to meet the needs of the commercial marketplace and are not well suited to the needs of cultural institutions. Obviously, enterprise level DAM solutions are quite expensive off the shelf, and they typically require substantial additional resources on the part of the buying organization to further customize and maintain their implementations. For cultural institutions of all sizes that may benefit from these DAM systems, the monetary and staff resources required to implement and maintain these systems is quite high. Commercial enterprises also have very different expectations about how long a DAM will serve their organizations. A span of several years using a particular DAM solution may meet a business’ expectations for return on investment, but if a cultural institution is willing to make these resource commitments, it expects to leverage that system for a longer period of time.
Likewise, within commercial enterprises, the relative value of data stored within a DAM may decrease rapidly over time, while the value of the digital assets held by cultural institutions is more difficult to quantify. Finally, commercial enterprises generally don’t have the same concept of “collections” of digital assets that a library or museum does and the data and metadata in commercial enterprises is generally much more simply structured. For example, businesses typically lack the notion of complex digital objects comprised of multiple datastreams, or hierarchically organized data, that are common in cultural institutions.
Cornell University Library (CUL), in collaboration with the Cornell College of Veterinary Medicine (CVM), Cornell Information Technologies (CIT) and Fedora Commons, is taking the lead in developing a generalized DAM system based on the open source Fedora repository software that directly addresses the challenges of managing digital assets produced within cultural institutions. Some of the complex issues that this project will address include:
- data ownership control and rights management,
- complex metadata requirements,
- data quality and integrity,
- data longevity, including stewardship into the future,
- repurposing data for both short- and long-term use, and
- ensuring that management of, and access to, digital assets promotes efficiency, productivity and profitability.
The DAM application will focus on issues of storage, retrieval and other aspects of managing digital data, including preservation. Public facing applications, such as learning management systems or Web based digital collections that reuse the data managed by the DAM for more specific purposes, will be built outside of the DAM to isolate data management from specific data applications. CUL developers have been influenced by the approach of the National Science Digital Library (NDSL) and its NDSL Data Repository (NDR) plug-ins that allow sophisticated integration between Fedora and tools already familiar to those working in cultural institutions, such as blogs and wikis, as well as learning management systems like Sakai and Moodle.
The project will proceed in three phases:
- CUL will work with identified collections of data and faculty members within CVM to develop a prototype to explore the identified issues in a pilot project.
- CUL will work with CIT to transform the prototype DAM application, further generalizing and scaling the software, to support its use as a centralized campus DAM service for faculty and staff running on top of a centralized Fedora hosting environment.
- CUL will work with Fedora Commons to explore further generalizing the DAM application to be a vertical market “solution bundle” that can distributed open source to other cultural institutions.
Although historically, DAM applications have developed largely independently of digital repositories, this project attempts to use the flexible nature of Fedora to create a digital infrastructure in which these two rich areas of development can compliment each other synergistically.