Art History and Computer Science are two disciplines with different approaches in doing research. Nevertheless, the digital material on art that is increasingly provided by museums or other cultural institutions with their online collections is interesting for both sides. While art historians have new possibilities to access, use, analyze and interpret their source material, computer scientists find datasets of artworks and thus images and metadata with quite particular characteristics. The challenge for art historians is to acquire the necessary technological knowledge and to apply suitable tools in order to draw upon the strengths a digital workflow offers. These are prerequisites for being able to adequately apply the advantages of technology to study art history in a digital environment and, at the same time, to establish new methods for the discipline. Computer scientists are interested in cultural data as a basis to develop applications and tools. Or they are curious to test it with existing prototypes to see what modifications are in need that these remain useful for that kind of data. This coincides with recent advances in computer science that particularly include topics in areas with a focus on visual data, like computer vision or machine learning with images, as well as all aspects around information retrieval and the semantic web.
Several interesting resources from cultural institutions exist such as the Rijksmuseum, the Tate, but also on Europeana that provide images of artworks from their collections or at least the corresponding metadata that can be reused. The Getty Vocabularies, e.g., the United List of Artist Names (ULAN), are a powerful resource to enrich the metadata provided by the museums. Such a record includes information about spelling variants of the artist’s name across different languages, her or his birth and death place, further biographical facts and relationships with other artists. This small overview alone makes clear that technologically far more is possible than just using the databases to find images through a search mask and use these to traditionally study the artwork, compare it with other examples and write a text about it. Instead, by using the power of the semantic web, networks can be created with computational methods and allow to get an overview about a specific data set, which ideally allows to get new insights, different interpretations and interesting conclusions. Numerical values, such as color, amounts of digital objects or information etc., can be processed easily with computer technology. This means that statistics become important, what is contrary to the usual methods in art history. To use the data provided in digital art collections, e.g., in a data visualization – good examples can be found on Google Arts & Culture Experiments –, it is necessary to have the ability of how to use these APIs.
There is often a gap between the knowledge and the skills of art historians on the one hand and computer scientists on the other. Whereas in art historical projects usually a satisfying use of technology is missing, innovative applications that use digital art collections are often interesting from a technological point of view and allow to predict future developments of, e.g., image search methods, but lack of providing in-depth knowledge of wider art historical relations or cultural contexts that are related to the artworks from different periods. It is important to take both sides equally into account for sophisticated tools or applications. Therefore, the idea of Coding Dürer is to offer art historians and information scientists a chance to collaborate in mixed groups to find challenging ideas and to initiate a fruitful dialog between both disciplines. The main focus is to enable the collaboration in interdisciplinary groups. This means that the largest part of the time is reserved to work on small, experimental projects with the aim to present a prototype at the end of the week. A few presentations by invited speakers during the week offer inputs from other perspectives. It will be also possible that participants give an insight in own projects that are of interest for the plenum.
