Small Archives, Big Problems

Digital Archiving: the Greeks had a word for it, "Sisyphean". In The Odyssey, Homer tells of Sisyphus, punished by the gods and forced to roll a large rock up a hill for eternity only to have it fall to the bottom, forcing Sisyphus to start out again on his endless task.

Sisyphus would nod in sage recognition were he to read "The Digital Dilemma", an extraordinary white paper issued in November 2007 by The Science and Technology Council of The Academy of Motion Pictures Arts and Sciences. As digital production has swept the entertainment business, the Council reports, digital archiving is beset by a lack of preparedness and solutions, what the report describes as the "absence of guaranteed long term access".

Among the conclusions documented: conventional analog archiving on film is a mature technology with relatively low cost ($25 a year to keep a feature film camera negative in cold storage), robust keeping properties (100 years minimum in a controlled passive environment) and a predictable outcome. Digital archiving involves fragile binary data that must be written to magneto-optical discs or data tape that is expensive to maintain, has limited keeping properties (5 years) and requires endless "active management" and "significant and perpetual spending" (1100% costlier than film). The limited keeping of digital data is due to both fragile media carriers and operating systems that change as frequently as car dealership showrooms.

If digital archiving is a dilemma of Homeric proportions for giant entertainment companies possessed of mandate, capital and resources, it's a real boogeyman for small archives racing to keep ahead of a vanishing past. Small archives have traditionally worked with small budgets that just manage to keep pace with analog archiving models. "Most seasoned media archivists have guarded praise for digital technology", says Dr. James V. D'Arc, curator at Brigham Young University's Motion Picture Archive at Provo, Utah. "We can distribute the digital content quickly, and we can accomplish repairs that could not be done with traditional methods. But in the short history of digital we have experienced alarming obsolescence or unfulfilled promise of performance. It's like a sucker punch in a boxing match" say D'Arc. "We don't know when the punch is coming, or from what direction. Media fails, operating systems change, and backward compatibility cannot be presumed."

D'Arc urges curators to keep their originals. "BYU remains committed to preserving the original artifact," he says. He points to their collection of composer Max Steiner's acetate recordings, created on Hollywood scoring stages decades ago. BYU re-recorded the fragile records to Digital Audio Tape in the 1990s. Within five years, the DATs failed catastrophically. Physical rust and digital dropouts made the preservation transfers unplayable. The archive had to return to the analog originals, still robust after nearly three-quarters of a century, and make new digital transfers.

While we await the advent of what the Academy's Tech Council calls a "rational digital preservation technology", for now the only sure path for digital archives is migration and redundancy: multiple copies, migrated frequently whether you're a large studio or a documentarian. Although the advent of electronic media systems seems to be a quick and a cheap way to capture, edit and distribute images, the preservation of this "fast food" media is quite expensive and requires constant tracking of content, technology and media degradation. The concern for the future is a significant loss of images where multiple copies migrated every 5 - 7 years has not happened and the cost to forensically recapture the data will be overwhelming to the content owner thus creating a black hole of content during this time of technology without longevity.