Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

May 27, 1999

Even in the information age, microfilm remains the best bet for

preserving library materials, prof argues

No digitization without preservation

Even in the information age, microfilm remains the best bet for preserving library materials, prof argues

"Digital documents last forever — or five years, whichever comes first."

–Jeff Rothenberg

Senior research scientist

RAND Corp.

The Clinton White House cranks out 6 million e-mail messages annually, then sends them to the National Archives for "permanent" storage — adding to the Archives' 4 billion-plus pieces of paper, photographs, maps and other holdings.

Closer to home, Pitt's University Library System (ULS) maintains 4 million books, subscribes to 25,000 journals and recently bought some breathing space by moving its growing archival collections to a new site in Point Breeze.

Obviously, not every book, document, article, music recording, etc. can be preserved forever.

But who decides what to keep and pitch? And in what format should archival materials be preserved?

One scholar who argues against the current trend of preserving library holdings by converting them to digital forms ("electronic" books and journals, CD-ROMs, etc.) is Sally Buchanan of Pitt's School of Information Sciences.

"If we're going to continue storing archival material in digital formats — and a growing amount of material is only published digitally — then we ultimately need to develop a universal [computer] language that will enable us to run original software on future computers without great expense," said Buchanan, who has consulted on library preservation projects around the world.

"Creating that language itself will be very expensive, and so far no organization has shown interest in developing it," she pointed out. "In the absence of a language that makes it easy to migrate old software, I don't think there's any preservation forma t we dare use except microfilm."

The shelf-life of microfilm produced on silver halide film and stored in a stable environment is 300-500 years. "Nobody likes to use it," Buchanan acknowledged, "but microfilm is still the best affordable archival storage medium we have. It's also a unive rsal master. You can make hard copies from it, microfilm, microfiche, even digital format copies."

Preserving books onto microfilm is less expensive than using digital technology, she added. "It costs $125 to microfilm an average-length book of, say, 300 pages. It costs $1 per page to scan and create a digital format for the same book, and the maintena nce costs of migrating it to new software are additional."

Buchanan said she sensed a disturbing trend while attending the American Library Association's January meeting in Philadelphia. "Several research library directors indicated they had decided microfilm is passe and preservation is going to turn increasingl y to digital formats," she said.

Some librarians recommended preserving "brittle" books — volumes printed on acid-based paper that cracks or crumbles with two foldings — by transferring them to digital form. "They're changing formats, but the second [digital] format is even less stable than the original, printed one," Buchanan complained.

"In my opinion, a big reason this is being done is because there's this craze for being seen as a cutting edge institution. Some librarians think: 'If digital formats are the cutting edge, then by God, that's what we're doing to do!'"

Buchanan commended a current ULS project to preserve 2,350 embrittled monographs from Pitt's renowned Bolivian Collection. Funded by a $219,000 grant from the National Endowment for the Humanities, the project will preserve onto microfilm documents that a re out-of-print, scarce, lightly damaged or stained. For a fee, ULS will make copies from microfilm and send them to purchasers anywhere in the world.

But Buchanan criticized a new collaboration between ULS and the Historical Society of Western Pennsylvania. The Historic Pittsburgh project is building a database of 19th-century and early 20th-century books and documents related to the region's history.

The project will enable students, professional scholars and amateur genealogists alike to search and view the materials — many of them fragile, one-of-a-kind texts — on their computer screens via ULS's Digital Research Library.

"Frankly, I think it's a mistake," Buchanan said.

"Don't get me wrong," she quickly added. "As an access tool, what Pitt and the Historical Society are doing is wonderful. People all over the United States and in other countries will gain access to historical material which has not yet been available to them unless they came to the Pitt campus.

"But much of this material is extremely fragile, and it's continuing to deteriorate. For preservationists like me, projects like this are problematic because you're spending lots of money digitizing data and yet nothing is being done to preserve the origi nal primary resource material.

"It's not just Pitt. Other libraries are doing similar things." In fact, Historic Pittsburgh is modeled after projects at the University of Michigan and the University of Virginia.

ULS Director Rush Miller said he is likewise wary of leaping to digital without considering preservation. That's one reason ULS maintains printed scholarly journals in addition to providing electronic versions. "Until publishers can guarantee archival acc ess to back issues electronically, and right now it's not in their financial interest to do so, we'll keep those old issues on our shelves," Miller said.

Converting library holdings to digital forms can open worlds of information to anyone with access to a computer and the Internet. (And in an effort to level the playing field between information age "haves" and "have-nots," a growing number of libraries , including the huge New York Public Library, provide free Internet access without even requiring users to flash a library card.)

But preservationists worry: How long will the information itself survive?

"There is as yet no viable long-term strategy to ensure that digital information will be readable in the future," RAND Corp. researcher Jeff Rothenberg warned in January in a report to the Council on Library and Information Resources (CLIR).

"Digital documents are vulnerable to loss via the decay and obsolescence of the media on which they are stored, and they become inaccessible and unreadable when the software needed to interpret them, or the hardware on which that software runs, becomes ob solete and is lost," Rothenberg wrote.

Preserving digital documents will require many millions of dollars, he concluded, because the problem extends beyond libraries to threaten government records (75 percent of which will be stored only in electronic form by the year 2000), scientific data, m edical records, e-commerce transactions and computer maps of toxic waste disposal sites, among other information.

The information superhighway is strewn with digital wrecks:

* In Oregon, a primary national database for people with disabilities recently vanished.

* In New York, a digital archive of Pennsylvania Railroad corporate records was accidently erased.

* Archivists at NASA's Jet Propulsion Laboratory estimate that vast amounts of NASA data stored on magnetic tape, including some transmissions from the 1976 Viking mission to Mars, have decomposed because nobody copied them or converted them to more stabl e formats.

Scholars labored unsuccessfully for 1,000 years to decipher ancient Egyptian hieroglyphics. Then one of Napoleon's officers uncovered the Rosetta Stone.

Digitally encoded data won't allow for serendipitous translations like that. Once the software and hardware no longer exist to break up digital data into 0s and 1s and assemble them in the proper binary code, the data is forever useless.

Buchanan shows her classes a videotape, produced by the American Film Foundation's Commission on Preservation and Access, that includes a chilling interview with the founder of a leading producer and publisher of electronic books and CD-ROMs.

"I can't imagine anything that we are making today will be available 100 years from now," Voyager Co. founder Robert Stein says in the interview. "I don't even think it will be available 25 years from now. In fact, a lot of stuff that we made seven or eig ht years ago can't be played" because of the fast turnover of computer hardware and software.

One irony of the information age is that the more sophisticated data storage technology becomes, the quicker it wears out or becomes obsolete.

Because of their relatively simple design, surviving models of Thomas A. Edison's phonographs and cylinder disks still reproduce audible, if scratchy, recordings of turn-of-the-century music and speeches. Vinyl LPs, while easily damaged by dust and heat, last for decades and it's still possible to find shops that repair stereo phonographs.

But CDs last only for 8-10 years before aural gaps begin appearing, except for special "archival copies" that are projected to remain pristine for 25-50 years — assuming machinery on which to play them survives that long.

Remember BETA video-cassette players? "Try finding spare parts for them today," Buchanan challenged.

Preservationists have proposed various labor-intensive methods of preserving digital information: maintaining old software and hardware in special museums; painstakingly "migrating" information to new software as older programs become obsolete; and the tr usty standby of printing hard copies of digital documents.

But the incredible glut of archival data works against those ideas, according to Buchanan, Rothenberg and other preservationists. They cite a 1996 study by the National Archives that indicated it would take 125 years, at current staff levels, to transfer the Archives' backlog of nontext materials such as photographs and videos onto more stable formats. Research libraries face similar, if smaller, challenges.

Buchanan said she had hoped Rothenberg's clarion calls to CLIR and in an earlier article in the journal Scientific American would inspire popular interest in preserving digital information. "But that hasn't happened yet," she said.

"We're at the same point now with digital and electronic formats that we were at the turn of the century with the 'brittle books' problem," Buchanan suggested. "Librarians began to raise the alarm about acidic paper and how quickly it was deteriorating in the late 19th century. But it wasn't until the 1970s, really, that the Library of Congress took a leadership role in raising awareness among top research libraries."

After Yale University's libraries found that 40 percent of their holdings were embrittled, and surveys at other libraries revealed similarly alarming conditions, Congress began allocating tens of millions of federal dollars annually to book preservation.< P> "I'm sorry to say that the demand to begin printing books on acid-free paper didn't come from the library profession itself," Buchanan said. "It came from the public, who said: 'What do you mean these books we're paying $75 for aren't going to last?' And it came from authors who went to their publishers and said: "If you don't print our work on permanent paper, we'll take our business elsewhere.'

"I think we're going to need that same kind of groundswell on behalf of [information preserved on] digital formats."

That groundswell, if it comes, will have to break through a wall of cultural indifference, according to Buchanan.

"Preservation, in general, is a hard-sell in North America. It probably has to do with our history of pushing west, always moving on to new frontiers. What's new is what's interesting. Fixing roads and bridges is boring.

"Mention library and archival preservation to Asians and Europeans, especially in the Scandinavian countries, and they're familiar with the problem and appreciative of the need to do something," Buchanan said. "Mention it to most Americans and Canadians, and they get glazed expressions." But Buchanan is no defeatist. "We can't just throw up our hands and say, 'It's hopeless!'

"The real issue is deliberate selection. Do we want to base what survives on fate or haphazard technological changes? Or should we, instead, think carefully about what we want to save, and let the rest go?

"That's the precedent we set with book preservation," Buchanan said. "It was clear we couldn't save every printed book, so librarians and archivists consulted with scholars from a wide variety of fields to systematically determine what should be preserved , and for how long. Six months? Six years? A century?

"Now we need to do the same thing for electronic formats."

— Bruce Steele


Leave a Reply