Which metadata for Ancient Arabic Manuscripts Cataloguing?

Three million ancient Arabic manuscripts are jealously preserved in libraries, in conservation institutions and in private homes. A non-consulted document is like a dead document—its fate is related to its consultation by users. The access to this kind of document is very difficult because of two main reasons: first, the fragile state of the manuscripts that make them easily degradable, necessarily limiting handling. Second, the dispersion of the manuscripts in different locations in the world makes access difficult. Thus, the digitization and the online availability of the ancient Arabic manuscripts are sought as a solution. The question remains: how to access the digitized images of the manuscripts? Despite the difficulty of interpreting the images, building the cataloguing is an imperative. Our paper deals with the Arabic manuscript cataloguing problems. It reviews the various aspects of the Arabic manuscripts cataloguing metadata. We propose a new cataloguing model and a potential interoperability project between catalogs encoded with different formats.