In his new book, Double Fold: Libraries And The Assault On Paper (Random House), Nicholson Baker brings up an interesting point. For thousands of years, paper records have allowed historians to glimpse human culture of the past. But scholars of the future, Baker points out, might not be so lucky — thanks in part to an over-reliance on technology.
What's gotten Baker worried is the microfilming of our libraries. In the 1970s, a movement began among librarians both in the United States and abroad to transfer key historic documents from paper to microfilm. Paper, it was argued, is impermanent, fragile and extremely susceptible to decay. Only through the then-current technology of microfilm could archivists effectively preserve these valuable historical records.
Ironically, in many cases the original paper copies were destroyed in the process. Bindings were often cut from books to allow the pages to lie flat for scanning. Newspapers were thrown away by the thousands — after all, libraries now had access to a "permanent" edition. In many cases, no further original copies of these publications are known to exist.
This destruction is a crime, says Baker, and one that has to stop. In his book, he attacks the studies conducted in the 1970s that "proved" microfilm aged more slowly than paper. The advertised 500-year lifespan, he claims, can occur only under the most meticulously maintained climate and temperature conditions — a task beyond the capabilities of many libraries.
These days, archivists no longer rely on microfilm to the extent they once did. Digital media is the rage among today's librarians. But Baker isn't too optimistic about that, either. Nor should he be.
We who are accustomed to using PCs in our daily lives tend to think of digital media as the perfect form of storage. Digital documents are infinitely reproducible, can be stored in very little space, and are easily transmitted. But that's not the same thing as permanence; far from it.
Just as microfilm was once said to last for centuries unchanged, similar longevity claims have been made about digital media. But these have largely yet to be proven. A CD-ROM is said to have an expected lifespan of about 500 years, yet how can we be sure? The CD format itself was only invented in 1984.
But even if its medium stands the test of time, digital documents can "die" in so many more ways than simple microfilm. For one thing, while an ancient Egyptian papyrus is "accessed" in much the same way as a modern book, with digital media there's no such standardization. Hardware, software and data encoding issues conspire to make reading older electronic files a tricky proposition.
For instance, what equipment should be used to record information digitally, once it's been compiled? CD-ROM? Floppy disk? Zip drive? Magnetic tape? And if so, which format? Each of these comes in a variety of flavors. And digital storage media becomes obsolete at an alarming rate.
Suppose I gave you a digital document on a piece of media that I told you was a Macintosh-format floppy disk. Do you think you could read the document from it? Athough every model on the current Macintosh product line is a direct descendant of the first model to roll off Apple's assembly line, modern Macs don't come with floppy drives.
And depending on when the disk was written, it could be the early 400KB format floppy or the later 800KB format. Modern floppy drives store almost twice as much; however, their fine-tuned heads often have difficulty reading these older types of disks. For add-on USB floppy drives and the drive mechanisms that ship with Windows PCs, reading data from a 400KB format drive is a physical impossibility.
All this to read a diskette written less than 20 years ago. And yet there were many other formats that came before the Mac. Just be thankful you're not trying to read the long-dead 8-inch floppy format — or even that of Apple's Lisa. Some models of that short-lived machine used a type of disk that no other manufacturer has ever supported.
But, for the sake of argument, let's assume that you could read the data from the disk. What format would it be in? A Word document? A JPEG image? An MP3? One company called Octavo, founded by John Warnock, even offers digitized editions of rare books in PDF form. (Not surprising, considering Warnock's other company, Adobe, invented it.)
Yet while they may be useful and currently popular, none of these formats even existed when the Mac was first introduced. If our document was created using discontinued software, there's no guarantee that any current application would support its file format.
So how do we get around this problem?
Project Gutenberg, which is dedicated to digitizing written documents, has opted for a conservative approach. All project documents are encoded as plain ASCII text. ASCII is a comparatively venerable format that was invented in the 1960s.
But even this solution has its problems. Most notably, as an American invention, ASCII isn't particularly useful for representing languages other than English. One emerging standard that addresses this issue is Unicode, which supports a wider range of characters — but is barely 10 years old.
No matter what format is used, text is never stored digitally as letters and characters, the way it's recorded in ink on paper. It must be converted to binary representation, which must then be decoded and displayed before the document can be viewed. None of these translation schemes has stood the test of time to compare with plain old paper. In fact, all of them were invented in the last quarter of the past century.
This isn't to suggest that we should return to paper and forget digital media. In fact, Baker doesn't argue against the use of microfilm in his book. He argues that libraries should create film copies of documents while preserving the originals.
Yet what about cases where no printed original exists?
So much of our documentation today exists solely in electronic form. Consider how email has replaced the handwritten letter for personal correspondence. While Ken Burns was able to create his television documentaries of American history by studying letters written over the past two centuries, today's electronic letters often leave no trace of their passing. Once read, the documents are often discarded, or else meet their ends when our hard drives get reformatted.
What about Web sites? Surely there exists no better cultural record of the last few years of the 20th century. And yet very little of the content on current Web sites ever existed in print. Personal home pages, chat rooms, discussion boards and blogs all trust their content to digital media.
Even those much-maligned databases corporations use to track consumer purchasing patterns would be a goldmine to historians of the future. But what will become of them once the plug is pulled on the machines that house them? Will there be any record of the data gathered?
And what of electronic music? A hundred years from now, will historians have access to a mechanism to play tunes that have never existed other than as digitally recorded tracks?
The New York Times doesn't seem to think so. In its recent time capsule project, it stored audio recordings on a phonograph record made of solid nickel. And written works were recorded on acid-free paper.
Journalists are fond of referring to this as the Information Age. But what will become of the information once the age has ended? Has our information technology produced nothing more than a record that's destined to vanish into dust, even more surely than the piles of moldering pages that librarians once insisted be scanned onto microfilm?
