Steve Lance, Office of the Deputy CIO — Communications
Author's note: Thanks to the Media Vault Program's Michael Ashley and Patrick McGrath who contributed additional material to this article.
Research and teaching in the 21st century has become critically dependent on digital media. Students and faculty at UC Berkeley increasingly expect to connect digitally to each other and to vast banks of the University's valuable resources. So, how is the campus dealing with this paradigm shift, and what is being done both to safeguard these digital resources and to make them more accessible to our community of scholars?
Michael Ashley earned his doctorate in Anthropology and has been a part of the campus community for nearly 10 years. With a background in using digital media to support scholarship, Ashley has experienced the paradigm shift's benefits and perils. "Because technology is now an integral part of research," he said, "attention and commitment is needed to protect field notes, photographs, and other data from bit rot [1], hard drive crashes, and obsolescence caused by evolving file formats.
"As an archaeologist, I look back thousands of years ago to people who transferred their knowledge on stone tablets, and those stone tablets still exist," Ashley said. "We do not have a 'digital cuneiform' equivalent. And even when files survive, accessing them can be impossible if the applications and hardware that created them have disappeared from use."
Ashley now works in IST and is the solutions architect for the Media Vault Program (MVP), a new initiative being conducted in partnership by a small yet diverse team of campus representatives [2]. The group, facilitated by IST, consists of representatives from 13 campus organizations, including Art History, Anthropology, the Library, and many of the natural history museums.
The first project within the MVP is a "proof of concept", which has allowed the team to explore short-term needs and service priorities, while demonstrating the need for longer term, comprehensive preservation and sharing services for scholarly content. "It's really amazing what our participants have been able to accomplish," Ashley added. "We now have about half a million assets in 'the Vault'. It's a great start."
For the Phoebe A. Hearst Museum of Anthropology, one of the Proof of Concept participants, Michael Black, the museum's information systems manager said, "this is an opportunity to increase focus on our founding purpose. What Phoebe Hearst intended when she gave the money to establish the museum was for it to be an educational resource for the people of California. So we can't just say this is for faculty and students and staff only. We really have to reach out."
With enough exhibit space to hold about one percent of the museum's collections, reaching out has been difficult. Digital space is "just as valuable these days for research as shelf-space in libraries," Black added. "Before digitization, many of the museum's assets — which include a contemporary bust of Plato, world-class California Ethnology and Egyptian collections, wax-cylinder recordings of Native Californians singing in vanished languages, and an 8-foot-tall bird-shaped coffin from West Africa — were unknown even to staff."
Through the MVP, the museum is planning to make digital versions of the vast majority of its assets freely and publicly available, to "make it as easy as possible for teachers, students, and the curious public to get the information they want, when they want it," said Black.
Patrick McGrath, the MVP program manager from IST, discussed the subject recently. "This is a challenge experienced across higher education and research the world over. It's very complex and potentially expensive, but the payoff is potentially huge. A year ago when we were thinking about how to make progress in this arena, we recognized a number of notable activities taking place across the University, within departments, and especially in the Library and Educational Technology Services. Unfortunately, few digital media services exist that are widely available to the campus. Rather than just write another white paper, we felt that exploration through implementation and developing compelling case studies would be a way to elicit further campus support."
Based on this approach, CIO Shel Waggener and David Greenbaum, director of IST—Data Services, co-sponsored the project.
"By definition, a proof of concept is one with a fixed timeframe and is limited in scope," McGrath said. "Ours focuses on the capabilities of digital asset management software delivered with infrastructure from the campus data center. This means we're able to ensure offsite backups for the digital collections as part of the content workflow [3].
"By leveraging shared services like asset management, data storage and backup, the MVP can free participants to focus on research and teaching with media rather than managing bytes. And by increasing the visibility of Berkeley's 'silos of brilliant work' and perhaps tying them together, it may further the University's prestige among potential students, faculty, and donors."
"Building 'a warehouse of data' is not the MVP's goal," Ashley said. Instead, Ashley said that a production-ready media vault is an interchange that would allow users to preserve and access archival data of any size safely and privately, electing which assets to make available, and to whom. Researchers could share empirical data with colleagues on campus and internationally. Teachers could share lectures, presentations, and collections with students. Museums could share digital exhibits with other researchers and the public. Ideally, content would flow to trusted repositories, but can as easily flow to peer institutions, as well as into other systems, such as bSpace, Flickr, and CalPhotos.
"But as it exists, the Media Vault platform is not yet a comprehensive solution to address this challenge," McGrath said. "Part of the Proof of Concept's objective is to determine what functions such a tool should deliver in the short term, as a community project, with input from the participating departments and a larger survey group. Only in the coming months, when the participating departments prepare their own observations, will an outline of the campus's needs become clear," he said.
Currently, the MVP team and their partners are working to build a business case to present to the Campus Technology Council (CTC), through the CIO, for a permanent Media Vault Program and a potential initial service offering. "If the program is adopted, first-generation service could begin in the next fiscal year," McGrath said.
For more information, see the Media Vault Program Proof of Concept website.
[1] Gradual physical degradation of CDs and disks which can render them unreadable. http://en.wikipedia.org/wiki/Bit_rot.
[2] MVP Departmental representatives
[3] Content workflow is the sequence of automated or manual steps a file entering the Media Vault might pass through. Patrick McGrath provided this example of a possible content workflow for a digital image:
An image is
[ iNews | Search | IST | UC Berkeley Computing | UC Berkeley ]
iNews: UC Berkeley information technology news channels
Copyright 2007, The Regents of the University of California