by Nicola Bozzi

Birte Christensen-Dalsgaard holds a Ph.D. in Theoretical Atomic Physics, but she has been working for media archiving institutions involved in digital preservation – like the Aarhus University Library and the Royal Library – for many years now. Even if digital archives don’t sound as complex as theoretical atomic physics, in her presentation Christensen-Dalsgaard showed us that running them involves some pretty complicated reasoning. Starting with the premise that an archive should provide the best possible version of an object, and an appropriate context to access it, Birte and her team have worked hard on algorithms and models to lay out cost-effective strategies.

First of all, archives need to provide a navigation structure, which has to be kept up to date. Christensen-Dalsgaard and her team have to make sure access and presentation are maintained, while user experience has to abide by the last web x.0 principles (currently they are implementing the semantic layer introduced by web3.0). In order to keep the costs down, one of the strategies they have employed has consisted in ESP games, where users are encouraged to insert complex metadata, that a computer couldn’t do on its own, while playing a relaxing online game. This way everybody wins: the institution doesn’t spend all its money on human labor and users have a little fun while helping to make the service better.

Apart from crowd-sourcing, other aspects that don’t immediately come to mind when thinking of digital preservation are energy efficiency and sustainability. In this regard Birte pointed out how green IT can make a difference, when electricity costs force you to come up with new models for distribution – for example moving the servers from Denmark to Norway, where electricity is not coal-powered, can be a good strategy.
More in general, by showing a series of slides documenting he Cost Model for Digital Preservation project – which conceptualizes the task by breaking down its process and listing its most cost-critical aspects – Christensen-Dalsgaard made clear how important the model itself is, and how cultural heritage institutions need to find efficient algorithms to optimize performance and dialogue with local (and international) officials.