By Nick Kapur
The Digital Archive of Japan’s 2011 Disasters, a project of the Reischauer Institute of Japanese Studies at Harvard University, is a free-to-use, open source, participatory digital repository of “born digital” records of the 2011 triple disasters (earthquake, tsunami, and nuclear meltdown) that occurred in northeastern Japan. The archive is equally intended for three broad groups of users: researchers; teachers/students in language, history/culture, and digital humanities courses; and disaster victims, as a site of commemoration and remembrance.
In constructing our archive, we faced fundamental questions regarding how to gather, store, and make searchable and accessible a wide variety of digital content, including tweets, websites, images, audio files, video, and pdf documents, how to use these items in constructing historical narratives, and what implications, if any, a focus on digital sources has for the practice of history as a discipline. In the process, it became ever more clear to us that these are questions that historians will increasingly be forced to grapple with going forward. Even in a nation like Japan, which is somewhat notorious for its dogged adherence to paper records, large amounts of government and official records relating to the disaster were “always already” digital and never found their way into printed form. Twitter also came into its own in Japan during the disasters, as a crucial source of news when regular institutional, governmental, and journalistic information gathering broke down. The most dramatic video footage of the tsunami came not from news cameras but from cell phone video captured at ground level, and after the disaster people poured their hearts out not into paper diaries but blogs and internet message boards.
Our archive can be considered part of a recent trend of “disaster archiving,” which is a subset of a larger trend of “current event archiving.” Both of these trends are in their infancy, so we were able to use our project as a testbed for a variety of innovative new features that combine traditional archiving of documents and traditional aspects of website design with new ideas that take advantage of the fact that both the platform and the archived items are exclusively digital. I would like to highlight three of these ideas here.
First, our archive aims to provide a single, seamless, integrated repository of contents from a wide variety of partner archives within a single user interface. In principle, we do not store any of the items in the archive ourselves – instead, users use a single search box to search items from the digital repositories of over 30 content partners, whose contents are seamlessly integrated into our interface using APIs. The only information we store on our servers (with a few exceptions) is the metadata that we have generated for these various items within our site, such as tags, descriptions, and translations (the word “archive” thus becomes somewhat of a misnomer, but since the site feels like an archive to the user, we have continued to embrace the term). Not storing the actual content ourselves means that the holdings in our archive are infinitely extensible, depending only on the numbers of partners we can find to work with, but also requires a significant investment in building and maintaining trustful relationships with said partners.
Second, our archive is the inaugural instance of a powerful new, real-time mapping platform called “MapD.” Developed by a Harvard masters student in his spare time, this platform uses parallel graphics processing units (GPUs), rather than the server’s central processing unit (CPU) to graphically render huge data sets onto interactive maps in real time—a huge breakthrough compared to the several minutes required by commercial platforms currently on the market. Users of our archive can search among all the archived items with geolocation data on a scrollable, zoomable map of the world, narrowing their search using a variety of geographical or chronological parameters.
Finally, we have sought to make our entire archive interface as participatory as possible. This includes allowing any logged-in user to submit any type of digital content from across the web for inclusion in the archive using either a browser-based “bookmarklet” or more traditional web submission forms. Users can also contribute metadata, including tags and translations, which remains on the site to help guide future users. In addition, users can build personalised “collections” of items found in the archive, creating a mini exhibition on a specific topic or them, which can then be made public and shared with other users. Most powerfully of all, users can arrange, rearrange, and mash up items in our archive and from across the web into richly hyperlinked and potentially non-linear multimedia presentations using our custom-built “presentation editor,” currently available as a public beta. As with collections, presentations can be made public, serving not only as a way to present one’s own research, but potentially remaining on the site for other users to explore, following their own path through the presentation to make their own discoveries.
It is these participatory aspects that we feel hold the greatest potential for revolutionising how history is preserved, researched, taught, and commemorated. In contrast to the traditional image of an archive where people come to the archive to passively view what is stored there, or at best actively extract a few bits of information but leave no record of their interaction with the archive, a participatory archive puts the user at the very center of the archive’s functioning, maintenance, and expansion, blurring the lines between content creators and consumers, past and present history, historian and historical subject, teacher and student, and so on.
We already saw many of these role inversions in action in a course we taught at Harvard centered on the archive. By fully embracing the participatory and collaborative aspects of the archive, we found that the course instructors learned as much from the students as vice versa, and that students learned as much from each other as from the instructors. And as students and instructors collaborated to build historical records of the 2011 disasters in Japan, these records were then preserved in the archive and themselves became part of the historical record.
Although this new kind of participatory archive is in its early stages and we are only just beginning to scratch the surface of what it might become in the future, at its greatest potential, the participatory archive holds out the promise of radically leveling the playing field between the elite guardians of knowledge and ordinary citizens of the world. It almost goes without saying that our project is entirely open source and open access, without any subscriptions or paywalls of any kind. Users can even contribute directly to the computer code that runs our platform using our public GitHub code repository.
In the future we hope to continue to expand the participatory aspects of our archive by incorporating gamification and crowdsourced content curation. Although we recognise that our archive is an early attempt at a participatory archive, we hope that future event archiving projects will take some of the ideas we have begun to explore and further improve and refine them.
Nick Kapur is a postdoctoral fellow in East Asian Digital Humanities at Harvard University and serves as Project Manager of the Digital Archive of Japan’s 2011 Disasters.