[tt] CHE: Archiving Writers' Work in the Age of E-Mail

[Premise Checker]

Archiving Writers' Work in the Age of E-Mail
The Chronicle of Higher Education, 9.4.10
http://chronicle.com/weekly/v55/i31/31a00102.htm

Digital preservation lets scholars learn more than ever about authors

By STEVE KOLOWICH

Leslie Morris is used to handling John Updike's personal effects. For 
decades, Mr. Updike had been sending a steady stream of manuscripts and 
papers to Harvard University's Houghton Library, where Ms. Morris serves 
as a curator.

But in late February, several weeks after the iconic writer died, some 
boxes arrived with unexpected contents: approximately 50 
three-and-a-halfand five-and-a-quarter-inch floppy disks — artifacts from 
late in the author's career when he, like many of his peers, began using a 
word processor.

The floppies have presented a bit of a problem. While relatively modern to 
Mr. Updike — who rose to prominence back when publishers were still using 
Linotype machines — the disks are outmoded and damage-prone by today's 
standards. Ms. Morris, who curates modern books and manuscripts, has 
carefully stored them alongside his papers in a temperature-controlled 
room in the library "until we have a procedure here at Harvard on how to 
handle these materials."

Harvard isn't the only university puzzling over new media from old — and 
not-so-old — masters. Emory University recently received four laptops, an 
external hard drive, and a Palm Treo personal digital assistant from 
Salman Rushdie. The University of Texas at Austin recently acquired a 
series of Zip disks and a laptop containing Norman Mailer's files.

"Once we learned how to preserve paper, we were good," says Naomi L. 
Nelson, interim director of the manuscript, archives, and rare-book 
library at Emory University's Robert W. Woodruff Library. "That really 
hasn't changed a lot. With computers it's a whole different ballgame."

Still, three things are becoming clear. First, these trappings of the 
digital age will transform the way libraries preserve and exhibit literary 
collections. Second, universities are going to have to spend money on new 
equipment and training for their archivists. And finally, scholars will be 
able to learn more about writers than they ever have before.

In With the Old

Personal computers and external storage devices have been around for more 
than a quarter-century, but only now, as the famous literary figures of 
the 20th century begin to pass away, are these technologies showing up on 
archivists' doorsteps.

According to Ms. Morris, the Updike papers will be the first in the 
Houghton catalog to have a "significant magnetic-media component," and she 
realizes that old floppy disks are just the tip of the iceberg. The great 
American novelists of the digital era — the ones who own BlackBerrys, use 
Gmail, Facebook, and Twitter, and compose only on computer screens — will 
soon begin shipping their hard drives off to university libraries.

What happens then is something much on the minds of Matthew G. 
Kirschenbaum and Douglas L. Reside. Both Mr. Kirschenbaum, associate 
director of the Maryland Institute for Technology in the Humanities at the 
University of Maryland, and Mr. Reside, an assistant director at the 
institute, possess the collection of skills that may eventually be 
required of all 21st-century curators. In addition to holding doctorates 
in English, they are computer experts.

The institute, located in an austere warren of offices in the basement of 
the university's McKeldin Library, houses a mix of sleek new machines and 
clunky old ones. An easy office-chair roll away from his newest computer 
sits Mr. Kirschenbaum's oldest one: a small, gray box known as the Apple 
II. Mr. Reside's office contains similar artifacts, including a Commodore 
64 gaming console.

Amid the institute's state-of-the-art machines, these ridiculous-looking 
antiques are stark reminders of how rapidly computer technology has 
evolved, producing one of the major challenges of preservation in the 
digital age: compatibility.

The problem with the dizzying pace of computer evolution is that new 
machines are often incapable of learning old tricks — even if the tricks 
are not really that old. For instance, most new computers don't come with 
floppy-disk drives. And while Harvard will surely procure machines that 
can safely read Mr. Updike's disk-based papers, what if those papers were 
trapped inside an even older storage device — say, something resembling a 
Commodore 64 game cartridge? Future archivists must have the skills to 
retrieve them.

Brave New World

Archivists must also know how to transfer their data to new machines, 
since old machines can survive for only so long before their circuits give 
out.

That, Ms. Nelson says, calls for people with intimate knowledge of how the 
new stuff works, plus the resourcefulness to retrofit modernity's round 
holes to accommodate antiquity's square pegs. "We're still going to need 
people who are experts in the history of the book, people who study 
handwriting, organize paper collections, handle obsolete video formats, 
traditional photography. ... We're going to do everything we've been 
doing, and then we're going to be doing this."

Ms. Nelson understands this better than most: While Mr. Updike's floppy 
disks at Harvard probably contain simple text documents, the digital 
devices Mr. Rushdie donated to Emory contain entire ecosystems of data.

Writers today do a lot more on computers than they used to, and modern 
devices hold a lot more information about their users than old ones did. 
The laptop (and now the mobile device) has become the locus of social life 
as correspondence has migrated from letters and phone calls to e-mail and 
text chatting. Recreational reading and research have also increasingly 
moved to the Web.

Since a laptop logs basically everything its user does, preserving these 
data environments will allow the scholars of the future unprecedented 
insight into the minds of literary geniuses. "It's basically like giving 
someone the keys to your house," says Mr. Kirschenbaum.

The influence of authors' environments on their writing has always 
interested scholars. Marcel Proust, for example, is known to have been 
heavily influenced by the paintings he surrounded himself with when he 
penned the novel Remembrance of Things Past, between 1909 and 1922. 
Imagine if Proust had been writing 100 years later, on a laptop: What else 
we might be able to learn about his creative process.

The implications for scholarship are tremendous, Mr. Kirschenbaum says. 
Take a great digital-era author: "You could potentially look at a browser 
history, see that he visited a particular Web site on a particular day and 
time," he says. "And then if you were to go into the draft of one of his 
manuscripts, you could see that draft was edited at a particular day and 
hour, and you could establish a connection between something he was 
looking at on the Web with something that he then wrote."

In some cases, computer forensics can even hint at an author's influences 
beyond the screen. Mr. Reside recently mined data from old equipment 
belonging to Jonathan Larson, the late composer and playwright who earned 
a Pulitzer Prize posthumously for the musical Rent. In an early draft, 
Larson had a character suggest that the moonlight coming through the 
window is really "fluorescent light from the Gap." In the final draft, the 
lyric was "Spike Lee shooting down the street."

"From the time stamp on the digital files," Mr. Reside says, "I learned 
that the lyric was changed in the spring of 1992 ... when, I believe, 
Spike Lee was shooting Malcolm X in New York City."

A Deluge of Data

That is really just scratching the surface. Imagine how mapping the 
content of an author's Facebook profile, MySpace page, Flickr account, or 
Twitter feed might help scholars dissect that author's life and letters. 
The social-media generation has developed a habit of casually volunteering 
biographical information. When the great authors of that generation 
emerge, scholars may be pleased to find plenty of fodder for study already 
on the public record.

But that is where things get tricky. Information that lives inside a 
writer's personal hardware — like the data on Mr. Updike's floppy disks or 
Mr. Rushdie's hard drives — may not have physical dimensions, but it is at 
least attached to a single device that is owned by somebody. "It's 
physically here," says Mr. Kirschenbaum, gesturing toward a shelf of Apple 
Classic computers, donated to the Maryland institute by the poet Deena 
Larsen. "I can wrap my arms around it."

Not so with e-mail and social-media content. These are not programs run on 
individual computers; they are Web-based services, hosted remotely by 
companies like Facebook and Google. The content exists in an ethereal mass 
of data known in information-technology circles as "the cloud." There, Mr. 
Kirschenbaum says, "you get into this wilderness of competing terms of 
service."

With more and more information being stored on the Web, it is no longer 
clear who owns what.

For example, in February, Facebook rewrote its terms of service to stake a 
claim on all content that users put on their profiles. After a backlash, 
the company hastily backed off and reiterated that users own their own 
profile content. But the case is a reminder of the fluidity and ambiguity 
of ownership laws in the dawning era of shared media.

"Consumers don't really know their rights here, and many are so wowed with 
the convenience that they aren't asking themselves the tough questions 
yet," says Susan E. Thomas, a digital archivist and the librarian at the 
University of Oxford's Bodleian Library. "Right now we can collect boxes 
from the attic, but if the family request a cloud service to transfer the 
archive of their loved one to the Bodleian Library, will that happen? We 
haven't tried it yet, so I can't tell you."

"That's sort of the brick wall that every archivist knows they're hurtling 
toward at 100 miles per hour," says Mr. Kirschenbaum.

No Manual

Many other questions also remain unanswered. For example, how much 
information is too much? A 20th-century author's personal papers might be 
of manageable quantity — say, what she was able to store in her attic. 
Digital storage, on the other hand, is cheap, easy, and virtually 
unlimited. Mining, sorting, and archiving every bit of data stored an 
author's computers could become a chore of paralyzing tedium and 
diminishing value.

At present, researchers are wary of discarding anything. "The work of an 
author over their entire lifetime is such a fraction of the space you have 
on a server hard disk, so there's no reason to throw any of that away," 
says Mr. Kirschenbaum. However, he added, unless scholars are able to find 
what they want in that sea of data, it is not worth archiving in the first 
place.

The good news is that as computers are logging more data, reference 
technology is growing more sophisticated. And Ms. Nelson suggests that the 
new tools for interacting with born-digital artifacts — including a wiki 
functionality that could allow researchers to annotate materials and share 
their insights with others — may not be too far away.

New tools and new training, however, mean new money. Richard Ovenden, 
associate director of Oxford's Bodleian Library, says the speed at which 
universities adopt digital curation may depend on their willingness to 
divert funds from more traditional areas. And that could be at a slower 
pace than the speed of technological invention itself.