The death and life of digital archives.

How to Keep Digital Archives From Disappearing 

How to Keep Digital Archives From Disappearing 

The citizen’s guide to the future.
Nov. 4 2015 7:30 AM
FROM SLATE, NEW AMERICA, AND ASU

The Death and Life of Digital Archives 

Preserving the present doesn’t always work in the future.

151103_FUTURE_KillBinLadenVideoGame
Kill Bin Laden.

Screenshot of Kill Bin Laden, September 11 Digital Archive

This article is part of Future Tense, a collaboration among Arizona State University, New America, and Slate. On Thursday, Nov. 12, Future Tense will host an event in Washington, D.C., on the future of the library. For more information and to RSVP, visit the New America website.

When I entered the Sept. 11 Digital Archive, I did not expect to murder Osama bin Laden. Nevertheless, I found myself with a handgun. After I agreed to “Shoot that worthless motherfucker,” I fired a bullet into his eye socket. In another animation, a noosed bin Laden teeters atop a wall-eyed camel with the cue “Hang Him!” In a bit of macabre symmetry, players may even weaponize box cutters and commercial airliners.

Advertisement

The Kill Bin Laden game is among hundreds of homespun Flash animations—like The Torture Chamber, Fry Osama Bin Laden, and Turban Shooting—that speak to the undercurrents of confusion, fear, and hate in the aftermath of Sept. 11. In spite of or perhaps because of their DIY quality, these unofficial records invite visitors into a grieving process that is simultaneously intimate, ugly, and elliptical.

The Sept. 11 Digital Archive is not as a product of top-down curation but a process of open-ended participation, composed of born-digital ephemera. It is less a catalog than a swamp—a site of diversity, growth, and resistance. Its very superabundance, some 70,000 online items, conspires to disorient. While visitors might begin on a path like the “September 11 Bearing Witness” Exhibition, the search bar—pinned to the top of every screen—entices them to get lost in the contents of the digital archive.

I use the catch-all term digital archive with some trepidation. Digital does not distinguish between digitized (e.g. scans, transcriptions, recordings) or so-called born-digital materials (websites, emails, and digital photos). The term archive is itself contested, as detailed in Trevor Owens’ aptly titled entry for the Library of Congress. Owens, who is the senior program officer at the Institute of Museum and Library, recently shared with me several projects that illuminate the eclecticism of approaches to digital archives. One of the earliest projects, the Rossetti Archive, digitizes the correspondence and manuscripts of 19th-century writer Dante Gabriel Rossetti, alongside contextual materials such as contemporary periodicals; the result is, in Owens’ words, a “hybrid” of a critical edition and literary archive. The Salman Rushdie Papers, meanwhile, allow researchers to access born-digital materials by loading an emulator of Rushdie’s Macintosh Performa. The Sept. 11 Digital Archive was the first to crowd-source the collection of born-digital items, and its struggle for sustainability reveals how such projects are untenable without the human and financial support of universities, philanthropic organizations, and federal grants.

The Sept. 11 Digital Archive germinated in early 2002, when a team of 14 faculty and staff at the City University of New York’s American Social History Project and George Mason University’s Roy Rosenzweig Center for History and New Media began collecting born-digital materials related to the terrorist attacks. Without today’s most ubiquitous online platforms—Facebook, Instagram, Twitter—the team was forced to experiment with curatorial approaches: They created a Spanish language version of their website; collaborated with the Museum of Chinese in America to videotape and translate interviews with Chinatown residents; and one of managing directors of the project, Fritz Umbach, used his Arabic background to capture Flash videos from websites across the Middle East. The team also found that they received surges of materials around anniversaries. By 2011, they had gathered more than 150,000 digital items from across the world.

Advertisement

Stories about digital projects tend to stop here: a celebration of a specific, quantifiable achievement. Celebrated projects, such as Hypercities, Mapping the Republic of Letters, and Whitman Archive, have become indispensable teaching and research tools. (I referenced case studies from the Republic of Letters in the first chapter of my dissertation, and I have used historical maps from Hypercities to contextualize Horatio Alger stories.) However, as along as conversations about digital projects end with a product, there is a tendency to overlook the processes that enable proponents to sustain—and improve—resources. The Sept. 11 Digital Archive reveals a problem unique to born-digital materials: Unlike print materials, preserving them requires continuous translation. Whereas a manuscript might fall into disrepair over centuries, a file format might go obsolete with a software update. And just as it is unfair to ask progenitors to support projects indefinitely, it is disingenuous to promote online resources as finished when they are imbricated in cycles of growth and decay. Without continuous financial support and a long-term sustainability plan, a digital project relies upon the benevolence—and ad hoc maintenance—of advocates. These zombie projects, capable of neither growing nor dying, do a disservice to their publics.

Despite a continuous stream of submissions, funding for the Sept. 11 Digital Archive has come in spits and spurts. The project began with a generous grant ($300,000) from the Alfred P. Sloan Foundation, which covered administration through 2003. For the next eight years, the burden fell on the universities. At CUNY, one of the original team members, Stephen Brier, described an ongoing struggle to obtain institutional support. In 2003, the team anticipated that the Library of Congress would take ownership of the project; however, while the LOC has used the corpus to perform stress-testing, it has not assumed responsibility for the site. Instead, a handful of staff at George Mason University continues to maintain the servers. While upgrading metadata pro bono may be a labor of love, it is not a long-term sustainability plan.

After applying for numerous grants, the team finally secured funding through a National Park Service “Saving America’s Treasures” grant in 2011. That federal grant ($156,000) was generous, but it was no free lunch. The team had to report all spending to the National Endowment for the Humanities, and, as a 1:1 cost share, George Mason University donated an equal amount of staffing. (No small feat given that most staff at the Center for History and New Media are funded by the university or grants that cannot be cost-shared). Even so, the grant enabled the team to perform much-belated upgrades: migrating the site to a more stable platform (Omeka), upgrading metadata to make items easier to search and sort, reopening the collection portal to submissions, and making more items in the archive (such as oral histories) Web-accessible. Before the upgrade, about one-third of the 150,000 digital items were accessible online. Since, the number has grown to more than half.

Lest we assume this is the happy ending, the team had exhausted the grant by 2014. They cannot apply for an extension because the funding no longer exists. The Sept. 11 Digital Archive was both the first and last digital project to receive the grant, which was terminated as one of many “tough choices” in the FY 2011 Budget. The NEH has also struggled with budget cuts. Thanks to a substantial endowment, which supports more than 100 such projects at George Mason University, the Sept. 11 Digital Archive will endure, though without continuous investment its contents will inevitably become illegible, one file type at a time.

When they first lent support to Sept. 11 Digital Archive, the Sloan Foundation challenged creators to consider “what historians fifty years from now would want to know about the September 11 event to … construct a full historical narrative of what transpired.” While that challenge informed the methodology of the Sept. 11 Digital Archive, the nature of digital tools might prevent future visitors from accessing materials. Even today, I had difficulty viewing some materials. In fact, the Kill Bin Laden game I described at the top of this piece would not play in my default Web browser. In the future, barriers to entry will likely increase.

This problem is not unique to the Sept. 11 Digital Archive. Rather, the Sept. 11 Digital Archive registers a challenge for both print and online repositories: They do not sustain themselves. Unless citizens and institutions invest in long-term sustainability plans, all but the most noteworthy (and likely safest) records are at risk. It does not have to be that way. It is a choice, and one upon which collective memory depends.