Assessing the Preservation Condition of Large and Heterogeneous Electronic Records Collections with Visualization
AbstractAs collections become larger in size, more complex in structure and increasingly diverse in composition, new approaches are needed to help curators assess digital files and make decisions about their long-term preservation. We present research on the use of interactive visualization to analyze file characterization information for the purpose of assessing the preservation condition of a vast collection of complex electronic records. The case study collection contains over 1,000,000 files of diverse formats arranged in varied record structures and record groups. The visualization application uses tree maps and a relational database management system (RDBMS) to represent the collection's arrangement and to show available characterization information at different levels of aggregation, classification and abstraction. Through this visualization interface curators can interact dynamically with the collections' characterization information to discover trends, as well as compare and contrast various file characteristics across the collection. Curators may select and weight the variables that they want to analyze. They can pursue analysis workflows that go from a high-level overview of the collection's preservation condition based on file format risks, to obtaining more detailed results about the condition of record groups and individual records. While there are various digital preservation planning tools available, to our knowledge none have been designed specifically to visually present assessment information across vast and complex collections. We present research to address the need for such a tool.
Copyright for papers and articles published in this journal is retained by the authors, with first publication rights granted to the University of Edinburgh. It is a condition of publication that authors license their paper or article under a Creative Commons Attribution 4.0 International (CC BY 4.0) licence.