Same Same But Different - Comparing Rendering Environments for Interactive Digital Objects

M. Guttenbrunner, J. Wieners, A. Rauber, M. Thaller:
"Same Same But Different - Comparing Rendering Environments for Interactive Digital Objects";
Vortrag: Euromed 2010, Limassol, Zypern; 08.11.2010 - 13.11.2010; in:"Digital Heritage", Springer -Verlag Berlin Heidelberg, 6426/2010 (2010), S. 140 - 152.

[ Publication Database ]


Digital cultural heritage in interactive form can take different shapes. It can be either in the form of interactive virtual representations of non-digital objects like buildings or nature, but also as born digital materials like interactive art and video games. To preserve these materials for a long term, we need to perform preservation actions on them. To check the validity of these actions, the original and the changed form have to be compared. While static information like images or text documents can be migrated to new formats, especially digital objects which are interactive have to be preserved using new rendering environments. Original objects and migrated representations can usually be characterized from their stored form and the results compared. If the object stays in its original form but the rendering environment changes, the characterization has to be performed on the results of the rendering.
Similar principles apply also to all static objects, as preservation always needs to focus on the intellectual object, rather than the stored representation, which always has to include the view path, requiring us to compare rendered forms.
In this paper we show how the results of rendering an object in different environments can be compared. We present a workflow
with three stages that supports the execution of digital objects in a rendering environment, the application of interactive actions in a
standardized way to ensure no deviations due to different interactions, and the application XCL Layout processor that extends the characterized screenshots of the rendering results by adding information about significant areas in the screenshot allowing us to
compare the rendering results. We present case studies on interactive fiction and a chess program that show that the approach is valid and that the rendering results can be successfully compared.