Modelling the Cost and Quality of Preservation Imaging and Archiving

Cultural heritage institutions, such as archives, libraries and museums, have a tradition of copying their collections to enhance access and support preservation. The general aim of preservation driven copying is to secure or protect valuable information threatened by deterioration, handling, theft, fire and other risks. In this PhD thesis it is examined how one may evaluate the long‐term costs and benefits to cultural
heritage institutions of different preservation strategies for digital copies. The investigated alternatives are preserving the copies in a digital repository, and printing the files out on microfilm and preserving them in a non‐digital repository. In order to obtain empirical data and to understand the decisive cost factors in preservation copying, a case study was set up in which degrading sheet‐film negatives were digitised. Requirements for image quality and metadata were investigated and specifications based on best practice and testing established. Also, the image quality parameters, which influence the long term preservation costs, were identified. In
addition, the suitability for preservation of different image file formats and compression algorithms was evaluated and requirements for printing out digital files on film determined. The LIFE Costing Model was used to analyse cost data from the case study and it was concluded that the model is suitable for calculating expenses of production and preservation of digital copies as well as film copies. However, several shortcomings of the model were also revealed: First of all, the LIFE Costing Model
does not always comply with the OAIS Reference Model, and it does not provide operational accounting principles. Moreover, the level of detail of the Bit Stream Preservation stage in the LIFE Costing Model is not sufficient to allow for consistent capturing of dependencies on e.g. volume and bit security. Finally, that part of the model, which estimates the cost of functional preservation, has not been verified on a
larger set of empirical cost data. In co‐operation with another project aiming to cost preservation of digital materials held by national cultural heritage institutions in Denmark, a study was undertaken to provide a generic cost model for digital preservation. The outcome of the study is an activity based cost model, which accounts for full economic costs. It is structured around the functional descriptions in the OAIS Reference Model. The cost model divides the OAIS functions in a hierarchy of cost critical activities and measurable components, which are implemented as formulas in a spreadsheet. So far the model has only been completed for activities relating to preservation planning and digital migrations. The model has been tested on cost data from actual migration projects and adjusted accordingly. Overall the study found that the suggested model provides a sound basis for costing digital migrations and that the model can be extended to cover the other functional areas within digital preservation. However, assessment of the model’s accuracy and precision remains outstanding issues and will require more testing against real‐time cost data. Furthermore, the model will need to be more flexible to allow for handling various degrees of compliance with preservation requirements. Finally, it was also evident that more work is required to strengthen the model’s ability to estimate future costs, especially in relation to migration complexity and frequency. The work to design a cost model highlighted the need for documenting underlying assumptions and
objectives of the model and pointed to the need for a more systematic evaluation of the quality of repositories and the perceived benefits that different preservation strategies may bring. This also relates to a conducted investigation of preservation requirements for a shared bit preservation system, which describes how institutions with OAIS compliant repositories can build a service model for a shared bit repository offering differentiated services for bit integrity (security), confidentiality and accessibility. Furthermore, it relates to an evaluation of different types of cost benefit analysis methods, which has also been executed. Overall the study has outlined a framework for assessment of costs and benefits of preservation strategies, which will hopefully facilitate decision‐making within cultural heritage institutions and thus result in a more efficient and appropriate consumption of resources. The thesis starts with an introduction to preservation copying and positions the subject within the
research field. It then outlines the project and defines the research aim. Next, the thesis summarises the papers and technical reports on which it is based on and relates the documents to the conducted research. It concludes by assembling the outcome of the research and points to future opportunities.