Content Sprints vs. Archival Marathons

by Julie Gable

Digital information is easily created, quickly accessed, cheaply disseminated and a potential treasure trove of knowledge. The trade-off for such versatility, however, is fragility.

Whether converted to digital format or "born digital," electronic files rely on application software and operating systems to enable their use. Newer software versions are not backwards compatible indefinitely, and digital storage media, no matter how stable or long-lived, cannot guarantee format fidelity for the items stored on them.

The Council on Library and Information Resources, Washington, predicts that future generations will know more about the Civil War than the Gulf War because records of the latter are largely digital. In the business world, regulated industries must routinely keep records for many decades to comply with specific mandates. The problem that curators, historians, archivists and business people face is that today's digital systems are sprinters, while history is a marathon.

The dilemma is especially acute within pharmaceuticals, where the FDA's mandate for electronic submission of new drug applications has forced the issue of long-term maintenance of electronic materials. Joy Wallace-Craig, R&D records analyst for Procter & Gamble Pharmaceuticals, echoes a colleague's observation that "We all know what our organizations spent in order to become Y2K compliant, and those funds pale in comparison to what we must spend to address [electronic preservation]."

This view is shared at Eastman Kodak, Lockheed Martin and AIIM International, the partners behind www.digitalpreservation.org, a portal designed to raise awareness of challenges and offer practical perspectives on digital preservation. "We think the market for digital preservation is in the tens of billions," says Andy Lawrence, Worldwide Channel Manager for Micrographic and Hybrid Products at Kodak. Richard Palarea, vice president of technology at AIIM concedes that digital preservation may not be on radar screens now, but he predicts it will be in the foreseeable future.

Much research on keeping electronic resources has come from academia and government. Solutions proffered to date have significant trade-offs. Writing electronic documents to PDF seemed like a reasonable possibility, given the product's encapsulation technique and the ubiquity of its viewer. Critics pointed out, however, that PDF is proprietary to Adobe (though many other companies now support it) and, like any other software product, may or may not exist far into the future. Another focus is the potential of XML as an open, available, license-free file format that combines content and descriptive information into one unit that would presumably be readable by future software.

The idea of migrating files forward - that is, routinely copying and reformatting old files using new software versions - has merit, too, but proves a costly alternative. Gartner research from 1998 indicates that the real costs for such efforts can exceed estimates by 300 percent to 500 percent.

Software emulation, an idea from The Rand Corporation's Jeff Rothenberg, is appealing in its simplicity. Writing emulators to mimic earlier software could, at the very least, make obsolete file formats readable and printable once more, though emulators, too, are platform dependent.

Microfilm offers a proven, long-term solution, but as of now is limited to writing scanned image files directly to film. Expect that to change. Kodak, for one, plans to enable writing files created electronically directly to film by the beginning of next year through its Kodak Archive Writer. This may present a viable alternative for industries in transition.

Julie Gable (juliegable@aol.com) CDIA, LIT, is an independent consultant based in Philadelphia.