Writing Samples

Layers of Meaning in Digital Archives

The migration to the digital world is well underway. With that migration, new ideas, new technology, and new ways of working are being created. Objects must be stored, retrieved, analyzed, and shared to support research in the humanities. The data from that research must also be stored, managed, and shared. The digital archive or content management system provides these functions.

It might be easy if all information originated in the digital domain. Humanities research spans human existence, before Turing and Babbage, before Gutenberg, to the earliest records of humanity available to us. Those records do not come to us in a digital form, but it would be ideal to have a digital representation—note that this is not a replication—available. Wear and tear on the objects (and traveling researchers) would be greatly reduced.

The digital form should attempt to represent the object completely, in ways that are useful to anyone who wishes to interact with it. Consider a text. The “final form” of the text is sufficient for many readers. Maintaining the original formatting may not be a concern, so a file that contains no more than a stream of characters without special encoding would suffice.¹ On the other hand, the scholar might be interested in examining different aspects of the document such as its physical appearance, the types of non-textual markings that are found, or the changes from erasures and corrections that can be observed. That requires one or more different representations with the additional detail encoded in some way.

Another consideration will be the handling of the metadata. This is the point where the scholar needs to understand the capabilities of the digital archive. For example, blogging software such as WordPress can store an object, and it can store metadata about that object. Its native metadata schema has not been developed with the researcher or archivist in mind, for example, the Dublin Core standards. This does not mean that WordPress (or another general-purpose content management system) is unsuitable for use within a digital humanities environment. It may be more suited for certain tasks that do not require as much academic rigor (such as a platform for discussion among researchers), or it may need additional software to provide the necessary features.

Those examples touch upon how a single type of object would be stored in a digital archive. The considerations grow exponentially when one considers that within the humanities many different types of objects or representations of those objects come under scrutiny. The digital archive has an impossible task. It must provide a way to store, manage, and transmit everything about anything. Since that is not possible, those implementing a digital archive should understand the limitations imposed by reality. An archive might be optimized for a particular type of record (Sanderson, et al.). It might have an organizational structure that is difficult to work with if the real-world organization doesn't fit the structure (Koenig and Mikeal). It could be capable of storing and linking different representations of the same object and easily making those available to the researcher (Staples, Wayland, and Payette). These are all examples of the capabilities and limitations of digital archive systems today.

Storing and cataloging the digital objects is the primary focus of a digital archive, but not the sole focus. To be useful, the objects must be made available to users. The user interface should facilitate the researcher's work, should not put up artificial roadblocks, and ideally should assist in creating meaning. Flexibility is key, because the interactions between the user and the object will be unique for every combination of user and object. There can be a different meaning that comes from a new encounter. Interacting with multiple objects in a serial, one-at-a-time fashion offers one perspective, but how a user arranges those same objects in one canvas opens a new dimension of interpretation (Price, Koontz, and Lovings). Tom Scheinfeldt reflects that sometimes we use tools to answer questions, other times to find new questions (Cohen and Scheinfeldt 56). The ideal archive will assist us in both tasks.

Coste asks, “What are digital texts made of?” (17) They are complex compositions that have active lives of their own. They are built, layer-by-layer, with each layer adding something to the final product. Using this layered analogy can help greatly with understanding the digital archive, especially when one keeps in mind that not all layers are obvious—the infrastructure and “ultrastructure” (Coste 5-6) are largely unseen, but as important to the digital text as paper and ink are to the physical.

The purpose of this annotated bibliography is to situate the discussion of digital records and digital archives within the digital humanities, to highlight issues of the academy in a digital world, and to open thinking for what a digital archive needs to accomplish for the practitioner in the digital humanities. Case studies illuminate some aspect of a digital archive. Seeing the decision points, compromises, and successes gives insight into how to build an archive that helps the work of the scholar.

¹ This intentionally ignores that an encoding representation such as ASCII or UTF-8 is needed to even get to this point. That detail is significant in the broader discussion of representation, but unnecessarily complicates the example.