The Cranach Digital ArchiveOctober 16, 2013 4:07 pm Server, Case Study, IIPMooViewer
Guest Post by Gunnar Heydenreich & Jörg Stahlmann, Cranach Digital Archive
The Cranach Digital Archive (cda) is an interdisciplinary collaborative research resource, providing access to art historical, technical and conservation information on paintings by Lucas Cranach (c.1472-1553), his sons and his workshop. The repository has been available online since January 2012 and currently provides information on more than 700 paintings including around 8000 images and documents from 92 contributing institutions, including major museums such as the Metropolitan Museum in New York, the National Gallery in London, the Getty Museum in Los Angeles, the Alte Pinakothek in Munich and many more. The archive contains over 400 infrared-reflectograms, 200 X-radiographs, numerous technical reports and a literature database with more than 2100 entries.
The project started in 2009 and is currently in its second phase (2012-2014). During this period, the Cranach Digital Archive aims to expand the existing network, to develop a shared infrastructure and to increase its content in order to build the foundations for an innovative, comprehensive and collaboratively produced repository of knowledge about Lucas Cranach and his workshop that will be significantly different from the traditional model of the single-author catalogue raisonné.
The Cranach Digital Archive is a joint initiative of the Stiftung Museum Kunstpalast in Düsseldorf and the Cologne Institute of Conservation Sciences / Cologne University of Applied Sciences in collaboration with nine founding partner institutions, 18 associate partners and many project contributors. The project is funded by the Andrew W. Mellon Foundation.
For the Cranach Digital Archive we were required to create an online web presence accessible to everybody which should include a digital archive connecting all collected information of an object and be capable of delivering high resolution zoomable images. In collaboration with the Düsseldorf Art Archive (d:kult), we modified the existing structure in the collection management system TMS to store information from Word documents and thumbnail link-references to PDFs and images. The task was then to unite the various existing structures, which included TMS, a lot of additional information in the form of static Word and PDF documents and a large number of images in various formats and sizes to create a user friendly web database.
The collection management software acts as a version control system giving us the possibility to work and test the data provided by our project partners and entered or generated by our team before publishing it. In order to update we export all data from TMS to XML format, generating 4 different documents with Overall, Conservation, Literature and Thesauri-Data, each of which can be up to 50MB in size and contain thousands of entities and attributions. An application written in Java parses the documents and inserts or updates the data in our database.
For our infrastructure we decided to use a locally hosted server, due to the fact that our data and images would create an immense amount of traffic if uploading was carried out via the internet. Our server system is based on Ubuntu (long term support version) and the Apache web server, though we are considering switching to Zend-Server for performance reasons. Also we decided to work with a standard open source database and of course with IIPImage.
The IIPImage server system allows us to view, navigate and zoom our high resolution images in real-time. With more than 8000 images currently available, the IIPImage server combined with tiled multi-resolution TIFF images is the perfect tool to comfortably and efficiently remotely browse the high resolution images within the digital archive. The images were batch converted using a shell script, using the free image processing system VIPS. Compared to most image processing libraries, VIPS requires little memory and runs quickly and efficiently, especially on parallel multi-processor machines.
For security reasons we are happy that IIPImage provides us with the possibility to place the source TIFF images outside of the web server’s document root. The images are streamed and can only be viewed via the IIPImage viewer. We are looking at ways to optimize this and to enrich the resource as well as to implement new tools.
by Gunnar Heydenreich & Jörg Stahlmann, Cranach Digital Archive