The Digital Image Archive of Medieval Music (DIAMM)

The Digital Image Archive of Medieval Music (DIAMM) was founded in 1998 by two scholars from Oxford (Margaret Bent) and Royal Holloway University of London (Andrew Wathey). Although inspired by the need to collect images for a facsimile publication, the project saw preservation as a principal aim of its activities, since the materials with which it was concerned were in danger of loss or deterioration.

A significant number of fragments of Medieval polyphonic music survive worldwide, mainly thanks to the fact that they were written on parchment, a material that was often considered more valuable than what was written on it. The re-use of parchment to strengthen bindings or even, after cleaning and refinishing, as a new writing surface, has ensured the survival of a significant corpus of medieval music, supported by the contents of music manuscripts which survive intact. The fragments are in danger from natural decay, damage hastened by medieval ‘vandalism’, and poor husbandry resulting from early attempts at restoration. The need to create a very high quality colour-accurate record of this repertory was urgent, and high-resolution digital imaging seemed to provide the solution.

The manuscripts were digitized directly, since the project established early on that the quality of images it could obtain this way were far superior to what could be gained by digitizing surrogates.

The quality of the images, and the digital medium had the secondary, and very significant benefit of providing a resource for digital restoration which, thanks to the exceptional quality of the images, yielded results far outstripping what had been hoped for at the outset. The techniques developed by DIAMM have been applied widely across other repertories and subjects up to (most recently) diaries written in pencil by Imogen Holst in the 20^th century.

The project obtained a PhaseOne PowerPhase digital scanning back, which was replaced by a PowerPhase FX, purchased with Mellon Foundation support, marking the beginning of the Foundation’s interest in the work of DIAMM. Maximum capture area emulates a 144 megapixel CCD (High-street digital cameras capture in the region of 9-11 megapixels), and file sizes at 24-bit capture are approx 240-280 MB.

Although in 1998 the use of online resources for academic study was in its infancy, very soon it became clear that a digital delivery system for DIAMM was going to be essential. The Project approached the Mellon Foundation to allow them to scope and then create a significant system for delivery not only of the highest-quality images, but also accompanying metadata.

Technical development was taken over by the Centre for Computing in the Humanities (CCH) at King’s College London, and the website (http://www.diamm.ac.uk) now offers a feature-rich environment not only for the study of the documents in the DIAMM collection, but also printed catalogues which are sources for manuscript metadata. Researchers at CCH have developed an annotation tool which is personalised for each registered user, enabling them to create private or publicly available annotations which attach to any image in the collection. ‘Secondary’ images can be viewed alongside to the main image (e.g. ultra-violet, watermark and multiple restored images) and all are delivered at highest resolution with virtually no time-delay via the image-viewer. The metadata database will cover all music manuscript sources from the medieval period, not only those in the image collection, and already represents a significant academic resource in its own right.

The Project has received funding approval for a further three years of development which will significantly enhance the content, allow us to initiate self-funding digitization projects to increase the collection of online images, consult with academic and technical colleagues to create the best possible online research enviroment, and exploit a collaboration with four other metadata content projects in three other countries in order to create a single distributed database that will serve all the projects. Among other things this will create a metadata standard for music manuscript description in collaboration with major collection-holders. A very significant advance is that the five collaborating projects will be using, for the first time in any musicological project, xml-encoded music representation that will allow musical symbols to be searchable, as well as allowing users to choose the manner of representation of the music.

DIAMM Presentation