Effective Tools for Digital Object Management
The organization of digital files during the development of digital collections is as important as the organization of physical objects on library shelves. Digital objects can be lost as quickly as they are created without an established system for file naming and digitization workflow, resulting in lost time and productivity. The UNT Digital Projects Unit (DPU) has developed several methods for the digital object management of book and image collections including, but not limited to, standardizing the process of object file naming and using logical folder organization matched to an internal Wiki. These methods, while unique to our lab, are based on simple principles that can be implemented at any institution.
A digitization workflow begins with the organization of the physical objects in a logical and traceable way so retrieving the correct item to digitize does not interfere with production. In an environment where multiple technicians are creating digital content from the same source material, it is key that everyone knows the current status of the project so material does not get scanned repeatedly or not get scanned at all. The DPU utilizes a combination of physical tags, numbered carts, and an internal Wiki allowing for a parallel work environment. The internal Wiki also matches folder hierarchies in the digital realm and provides a layer of redundancy for when — not if — a file is misplaced.
Before the first digital object is created it is important to have identifiers assigned to each object. A unique and persistent identifier is used throughout the digitization process for file naming, structuring folders, and linking metadata records with digital objects. Upon digitization, the digital object is moved through a series of folders stacked in order of process from 1 to 7:
ToQC ToDekew ToResize ToOCR ToMetadata ToUpload Uploaded Ordering files into folders based upon the action needing to be applied allows more time to be spent on processing the files than determining what needs to be done next.
Additionally, this workflow is broad enough to allow for the multifarious projects undertaken by the DPU. Books require different handling than image collections. Photographic image collections can be more straightforward in their management as digital objects as each physical photograph usually has only one digital constituent (or two if you scan the back of the photo), but when a book is digitized many images are created. The DPU uses MagickNumbering, an internal naming schema for books that logically handles both object-order and pagination, while offering a robust quality control method. Through the use of MagickNumbers, an entire book can be managed in a single folder as master TIFF files.
By applying these methods the DPU is able to work on multiple projects at a given time while keeping a high level of organization, quality control, and output. The methods mentioned are also scalable from a project consisting of only a handful of photographs to a 100-year run of yearbooks to 10,000 negatives.