Introduction During the last decades, a large amount of information in cultural heritage institutions have been digitised, creating the basis for many different usage scenarios. We have been working in this area for the last 15 years, through projects such as the Museum Project in Norway (Holmen et al., 2004). We have developed routines, standardised methods and software for digitisation, collection management, research and education. In this paper, we will discuss long term sustainability of digital cultural heritage information. We will discuss the creation of sustainable digital collections, as well as some problems we have experienced in this process. We have divided the description of sustainability in three parts. First we will describe briefl y the technical part of sustainability work (section 2). After all, this is a well known research area on its own, and solutions to many of the problems at hand are known, although they may be hard to implement. We will then use the main part of the paper to discuss what we call organisational sustainability (section 3), which may be even more important than the technical part in the future — in our opinion, it may also be more diffi cult to solve. Finally, we briefl y address the scholarly part of sustainability (section 4). Technical Sustainability Technical sustainability is divided in two parts: preservation of the digital bit patterns and the ability to interpret the bit pattern according to the original intention. This is an area where important work is being done by international bodies such as UNESCO (2003), as well as national organisations such as the Library of Congress in the USA (2007), and the Digital Preservation Coalition in the UK (DPC,2001). It is evident that the use of open, transparent formats make sit easier to use preserved digital content. In this respect XML encoding is better compared to proprietary word processor formats, and uncompressed TIFF is more transparent than company-developed compressed image formats. In a museum context, though, there is often a need to store advanced reproductions of objects and sites, and there have been problems to fi nd open formats able to represent, in full, content exported from proprietary software packages. An example of this is CAD systems, where the open format SVG does not have the same expressive power as the proprietary DXF format (Westcott, 2005, p.6). It is generally a problem for applications using new formats, especially when they are heavily dependent upon presentation. Organisational Sustainability Although there is no sustainability without the technical part, described above, taken care of, the technical part alone is not enough. The organisation of the institution responsible for the information also has to be taken into consideration. In an information system for memory institutions it is important to store the history of the information. Digital versions of analogue sources should be stored as accurate replica, with new information linked to this set of historical data so that one always has access to up-to-date versions of the information, as well as to historical stages in the development of the information (Holmenetal.,2004, p.223). To actually store the fi les, the most important necessity is large, stable organisations taking responsibility. If the responsible institution is closed without a correct transfer of custody for the digital material, it can be lost easily. An example of this is the Newham Archive (Dunning, 2001) incident. When the Newham Museum Archaeological Service was closed down, only a quick and responsible reaction of the sacked staff saved the result of ten years of work in the form of a data dump on fl oppies. Even when the data fi les are kept, lack of necessary metadata may render them hard to interpret. In the Newham case the data were physically saved but a lot of work was needed to read the old data formats, and some of the information was not recoverable. Similar situations may even occur in large, stable organisations. The Bryggen Museum in Bergen, Norway, is a part of the University Museum in Bergen and documents the large excavation of the medieval town at the harbour in Bergen which took place from the 1950s to the 1970s. The museum stored the information in a large database. Eventually the system became obsolete and the database fi les were stored at the University. But there were no explicit routines for the packing and future unpacking of such digital information. Later, when the fi les were imported into a new system, parts of the original information were not recovered. Fortunately all the excavation documentation was originally done on paper so in principle no information was lost. Such incidents are not uncommon in the museum world. A general problem, present in both examples above, is the lack of metadata. The scope of each database table and column is well known when a system is developed, but if it is not documented, such meta-information is lost. In all sectors there is a movement away from paper to born digital information. When born digital data based on archaeological excavations is messed up or lost – and we are afraid this will happen – then parts of our cultural heritage are lost forever. An archaeological excavation destroys its own sources and an excavation cannot be repeated. For many current excavation projects a loss of data like the Bryggen Museum incident would have been a real catastrophe. The Newham example demonstrates weak planning for negative external effects on information sustainability, whereas the Bergen example shows how a lack of proper organisational responsibility for digital information may result in severe information loss. It is our impression that in many memory institutions there is too little competence on how to introduce information technologies in an organisation to secure both interchange of information between different parts of the organisation and long-term sustainability of the digital information. A general lack of strategies for long term preservation is documented in a recent Norwegian report (Gausdal, 2006, p.23). When plans are made in order to introduce new technology and information systems into an organisation one has to adapt the system to the organisation or the organisation to the system. This is often neglected and the information systems are not integrated in the everyday work of the staff. Thus, the best way to success is to do this in collaboration and understanding with the employees. This was pointed out by Professor Kristen Nygaard. In a paper published in 1992 describing the uptake of Simula I from 1965 onwards, Nygaard states: “It was evident that the Simula-based analyses were going to have a strong infl uence on the working conditions of the employees: job content, work intensity and rhythm, social cooperation patterns were typical examples” (Nygaard, 1992, p. 53). Nygaard focused on the situation in the ship building industry, which may be somewhat distant from the memory institutions. Mutate mutandis, the human mechanisms are the same. There is always a risk of persons in the organisation sabotaging or neglecting new systems. Scholarly Sustainability When a research project is fi nished, many researchers see the report or articles produced as the only output, and are confi dent that the library will take care of their preservation. But research in the humanities and beyond are often based on material collected by the researcher, such as ethnographic objects, sound recordings, images, and notes. The scholarly conclusions are then based on such sources. To sustain links from sources to testable conclusions, they have to be stored so that they are accessible to future researchers. But even in museums, this is often done only in a partial manner. Objects may fi nd their way into the collections. But images, recordings and notes are often seen as the researcher’s private property and responsibility, and may be lost when her career is ended. Examples of this are hard to document, though, because such decisions are not generally made public. Conclusion Sustainability of data in the cultural heritage sector is, as we have seen, not just a technical challenge. The sustainability is eased by the use of open and transparent standards. It is necessary to ensure the existence of well funded permanent organisation like national archives and libraries. Datasets from museums are often not considered to lie within the preservation scope of the existing organisations. Either this has to be changed or the large museums have to act as digital archives of museum data in general. However, the most important measure to ensure sustainability is to increase the awareness of the challenge among curators and scholars. If not, large amounts of irreplaceable research documentation will continue to be lost. References DPC(2001): “Digital preservation coalition”. http://www. dpconline.org/graphics. Dunning, Alastair(2001): “Excavating data: Retrieving the Newham archive”. http://ahds.ac.uk/creating/case-studies/ newham/. Gausdal, Ranveig Låg (editor) (2006): Cultural heritage for all — on digitisation, digital preservation and digital dissemination in the archive, library and museum sector. A report by the Working Group on Digitisation, the Norwegian Digital Library. ABM-Utvikling. Holmen, Jon; Ore, Christian-Emil and Eide,Øyvind(2004): “Documenting two histories at once: Digging into archaeology”. In: Enter the Past. The E-way into the Four Dimensions of Cultural Heritage. BAR, BAR International Series 1227, pp. 221-224 LC(2007): “The library of congress. Digital preservation. http://www.digitalpreservation.gov. Nygaard, Kristen(1992): “How many choices do we make? How many are diffi cult? In: Software Development and Reality Construction, edited by Floyd C., Züllighoven H., Budde R., and R., Keil-Slawik. Springer-Verlag, Berlin, pp. 52-59. UNESCO (2003): “Charter on the preservation of digital heritage. Adopted at the 32nd session of the 9/10 general conference of UNESCO” Technical Report, UNESCO. http:// portal.unesco.org/ci/en/fi les/13367/10700115911Charter_ en.pdf/Charter_en.pdf. Westcott, Keith(2005): Preservation Handbook. Computer Aided Design (CAD). Arts and Humanities Data Service. http:// ahds.ac.uk/preservation/cad-preservation-handbook.pdf.
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.