Tools for Preservation and Use of Complex and Diverse Digital Resources

To preserve digitally encoded information over the long term following the OAIS Reference Model requires that this information remains accessible, understandable and usable by a specified Designated Community. These are significant challenges for repositories, particularly when dealing with scientific data where the semantics must be associated with the data so that one can answer the basic questions such as “what does this number actually measure?”. Answering these basic questions lead to techniques which fit naturally into a strategy for re-use of information across disciplines. The CASPAR project (http://www.casparpreserves.eu) is an EU part funded project with total spend of 16MEuros which is trying to faithfully implement almost all aspects of the OAIS Reference Model in particular the Information Model. The latter involves tools for capturing all types of Representation Information (Structure, Semantics and all Other types), and tools for defining the Designated Community. This paper will describe the tools and infrastructure components which have been implemented by the CASPAR project to support repositories in their task of long term preservation of digital resources. We address also the capture and preservation of digital rights management and evidence of authenticity associated with digital objects. Moreover examples of ways to evaluate a variety of preservation strategies will be discussed as will examples of integrating the use of these infrastructure components and tools into existing repository systems. Examples will be given of a rich selection of digital objects which encode information from a variety of disciplines including science, cultural heritage and also contemporary performing arts. While there are many common aspects this selection also provides a rich set of distinct challenges both technically as well as organisationally. We will compare other preservation techniques which are in common use and show their deficiencies when dealing with these challenges. For example, even a relatively simple document may be a container of separate objects which have their own, say, provenance, and also have important and complex relationships to other objects both within and outside the containing document. We will demonstrate techniques for dealing with this and the many other challenges and compare them with other types of solutions from other projects.