The following is a guest post by Jefferson Bailey, Fellow at the Library of Congress’s Office of Strategic Initiatives.
A number of us around the office have fielded some interesting questions recently, both at public events and over email, regarding digital preservation’s susceptibility to what I will call, for lack of a better term, cataclysmic occurrences.
Electrical grid failures, mega earthquakes, solar storms, and electro-magnetic pulses have all been voiced as providing a distinct and inarguable counterpoint to our exhortations about the need for, and importance of, digital preservation. Whether prompted by the social turbulence that often attends technological change or simply a product of beliefs about the year 2012, there has been a noted uptick in the voiced suspicion of digital preservation’s viability in times of, well, calamity.
But instead of dismissing such comments, I was compelled to consider the unique characteristics of digital information that would prompt these rejoinders. I can’t imagine that book and paper conservators are often asked how we will be able to preserve print collections during a catastrophic shortage of bone folders or a plague of acidic zombies. So what is it about digital content that incites this type of rebuttal?
There are two characteristics of digital objects, I think, that underlie these claims. Both are characteristics that in a more nuanced sense do account for many of the challenges inherent in preserving digital materials. These two qualities are a digital object’s dependencies and its opacities.
Obviously, many of these apocalyptic comments originate from computational machinery’s functional dependence on the power grid. No argument there, though running fixity checks on my digital content would be low on my list of concerns in the event of a complete and indefinite worldwide power failure. But energy dependence is fundamentally not so different a concern than other dependencies that are issues in digital preservation, specifically hardware and software dependencies.
Hardware dependency refers to the fact that accessing digital content on storage media requires specific pieces of hardware – and that hardware can have a very short lifespan both commercially and operationally. Just as we may need the technology of reading glasses to read books, we need a specific drive to play a 5” floppy disk or a USB port to access a thumb drive.
FREDs and catweasels are one attempt to address this problem. Digital objects are also dependent upon the fragility of storage media. A scratched vinyl record may only add an occasional click or pop but a scratched DVD can be entirely inaccessible.
Software, like hardware, also tends to be of a limited lifespan. Operating systems often can’t identify or open unrecognized file formats and formats themselves require their own sustainability and preservation efforts. Because of these issues, one can begin to understand how digital content’s manifold dependencies can be exaggerated into a more extreme belief in its broader vulnerability in times of disaster.
A digital object’s perceived intangibility may also play a part in presumptions of susceptibility. It can be hard for us to comprehend digital objects’ physical appearance or location, as storage is often remote or unseen. Digital objects have a material presence on a spinning disk, but how many of us have seen the actual disk of our hard drive?
When we think of a digital object’s appearance, we think of how that object looks on our monitors and screens, even though its display is only the final stage of multiple mechanical, magnetic, and translational events responsible for creating that representation. The disassociation between the functions of hardware and the on-screen interface has been referred to as “system opacity” by John Seely Brown. Much as storage in the cloud compounds the abstractness of a digital object’s location, the scintillating thinness of its display on a screen plays a part in calling into question its durability.
That we think of digital objects primarily as forms within glowing windows is a characteristic that Matt Kirschenbaum has call our “medial ideology” and in other contexts have been referred to as “screen essentialism.” If it is not on our screens, we wonder, does it even exist at all? The disjunction and confusion that arise from this interplay between storage and location as abstraction and illuminated representation as corporeality no doubt compels many to think of digital objects as uniquely ephemeral or without substance – in times both good and dire.
Conspiracy theories are often a corruption or embellishment of existing social or cultural currents. In a similar sense, the recent comments we have received about digital preservation’s vulnerabilities may seem outlandish on the surface, but in a way they are predicated upon many of the properties of digital material that vex preservationists on a practical level. Of course, focusing on the fallibilities of digital objects overlooks their many positive and transformative qualities that facilitate, not complicate, preservation. But that is for another blog post.
I don’t doubt that we will continue to field the occasional question about the meaning and import of digital preservation during the end of days. But in the long run, regardless of whether digital preservation practices can survive the invasion of giant space ants, we will continue to argue its essential place in preserving our culture and knowledge — even as I for one welcome our new insect overlords.
Updated 5/18/2012: fixed broken link.