Rushdie's Computers: Born-Digital Archives and Humanities Scholarship

As questions and concerns about digital preservation and
sustainability become increasingly audible in the spheres
of digital humanities and humanities computing, the necessity
to build strong ties between digital humanities and digital
libraries only intensifies. Howard Besser, in his essay, The Past,
Present, and Future of Digital Libraries,” underscores the
connections between these two fields. He explains digital
libraries not only “provide access to original source material,
contextualization, and commentaries, but they also provide a
set of additional resources and service”.1 Besser then delineates
some of these resources and services, including digital
collections of traditional print materials, lexical analysis, and
increased accessibility. In addition to these important
contributions to humanities research, I would like to highlight
the emerging role libraries play in processing, preserving,
maintaining, and providing access to important archives that
house born-digital content. This role will not only contribute
to humanities research in decades to come, but will also impact
how research is performed by directing what content is made
available and how researches may access it.
In this paper, I will examine the Emory University Libraries’
acquisition and processing of a singular personal archive as a
case study to explore the methods and practices of handling
born-digital archival materials and the implications such
methodologies and their outcomes may have on humanities
research.
Emory University’s acquisition of Salman Rushdie’s personal
archive represents an important addition to the Manuscript and
Rare Books Library, and contributes significantly to the
University’s digital library resources and research. Rushdie’s
rich personal archive includes traditional manuscript materials
such as journals, personal correspondence, and notebooks, as
well as less traditional archival materials, namely a series of
personal computers that cover a significant span in his personal
and literary life. This digital archive includes five computers,
one early Macintosh desktop and four Macintosh laptops,
including both obsolete and current models. While MARBL
has previously acquired collections containing some digital
materials, Rushdie’s computers represent the first significant, sizeable digital component to the University’s extensive
holdings of rare and unique materials.
Such an acquisition requires archivists to engage with
technologists to ensure that the library can most effectively
serve current and future researchers and scholars. The curation
of such an archive raises important questions about how libraries
should process, index, and present these materials while
simultaneously addressing preservation and authenticity
concerns. Such questions include: What is the research value
of such an archive? How important is the physical artifact? Do
researchers need access to exact systems emulation? Is
providing search and browse access to the data sufficient or
will researchers be interested in Rushdie’s original directory
structure? Once data is migrated from the original environments,
do we continue to maintain those outdated systems? How do
archives sustain both master and access instances of born-digital
archives?
As digital librarians and archivists at Emory begin processing
the born-digital components of this important archive, they
must keep these questions, and the host of secondary concerns
circulating around them, in the foreground of workflow and
process discussions. In this paper, I will argue for the
importance of balancing the urgent needs of data and system
stabilization with the more long-term challenges of considering
the ideal outcomes and products of processing and providing
access to a rare and unique born-digital archive. This talk will
track the early stages of processing the physical and digital
materials comprising Rushdie’s digital archive, outline
approaches to handling the more complex processing
requirements, discuss proposed approaches to presenting the
archive to both local and distant researchers, and generalize
observations drawn from the experiences with this born-digital
archive to broader implications within digital libraries, digital
curation, and humanities computing.
With the acquisition of any archive, a research library takes on
multiple responsibilities to preserve, index, and provide access
to the rare and unique materials. Libraries—digital, brick and
mortar, or otherwise—must “incorporate the component of
stewardship over a collection.”2 Such stewardship carries with
it important responsibilities, especially in cases of archival
materials. Thus, with the arrival of the first shipment of
Rushdie’s digital archive, consisting of three out of the five
computers, our library and archive staff faced both immediate
preservation demands and distant challenges for archival
curation. We elected to produce a workflow that is both staged
and modular, which I will only summarize here. The first task
is to provide a secure and stable environment for the machines
themselves. As our archives had not previously included
technological artifacts, this first step required some intensive
space and environment analysis. Once we stabilized the physical
objects, this born digital archive next challenged staff with
questions of data recovery, data preservation, and data
duplication. Such challenges prompted us to develop a
partitioned data architecture that duplicates and preserves all
master data. The original is preserved, a master duplicate is
generated and stored darkly, a duplicate collection of the master
database is housed in a secure repository for in-house processing
and staging, and, finally, a fully-processed instance of the data
is made available through a production database. Such an
approach provides for preservation of original artifacts and
master data, while ensuring a level of security for data while it
is being processed for embargoed material.
In addition to preservation and security, authenticity of the
archived data is of particular importance to archivists, digital
librarians, and humanities researchers. Graham Barwell’s
discussions about originality and authenticity within the fields
of textual studies and electronic textuality resonate with archives
such as Rushdie’s born-digital materials.3 How can we most
authentically represent the digital archives included in the
Rushdie collection? Is the data the only component of real
research value or is the context that holds this data, the
paratextual elements, if you will, of equal importance to
researchers? I will explore these questions and provide
illustrations from our processing of the Rushdie archive to offer
some preliminary insights.
1. From Besser’s essay, “The Past, Present, and Future of Digital
Libraries” in A Companion to Digital Humanities, 2004, p. 557.
2. Besser, p. 559.
3. Graham Barwell, “Original, Authentic, Copy: Conceptual Issues
in Digital Texts” in Literary & Linguistic Computing 20.4 (1995)
see pages 416, 418, and 419.