Conservation and collection management: Integration or isolation

Abstract

The reason for documenting conservation work has been described by Organ’ as falling into three categories: administrative, ethical and practical. The administrative category deals with the management of objects, logging them in, and around a conservation department, assigning conservators to work on them and calculating the resources spent in their examination and treatment. The ethical category involves the requirement for conservators to examine and record both the condition of an object before treatment and the details of any conservation-materials or procedures which may be carried out. The third distinct category, the practical, involves the retrieval of information about the success or failure of different methods of treatment. It is because of this practical need to link a number of variables that automated methods in conservation recording are required.* We may be looking to tie down a number of factors: for example, which copper alloy artifacts from the Cyclades, dated to the Mycenean Bronze Age have a composition of more than five per cent arsenic and have required more than three repeated stabilizations with benzotriazole at a concentration of more than fifteen per cent in methanol? This is the type of query that can only be handled satisfactorily by a computer-held database. To the three categories listed above we can add a fourth-the large area of supporting information needed to inform conservators as to the best, most up-to-date methods for treatment. This information can be found by searching through the relevant conservation literature or a conservation-material index. The task is becoming much easier with the introduction of international on-line databases such as the Conservation Information Network.3 The tendency has been to look at conservation documentation as distinct from the recording carried out in other areas of an institution. As we crystallize our ideas on conservation recording, large shared databases are being designed which will hold all the information relevant to the examination and treatment of objects. However, other information held about objects is of equal interest, and the aim should be to build up a total history of any given object. The rationale for automated information retrieval depends on the possibility of shared information. A database involves a pool of data which can be looked at in different ways by different people and looked at differently over time. Database design should start with a definition of an organization’s complete data resources, key interrelationships between individual elements of data are identified and the pieces put together as an integrated whole rather than as a collection of separate and independent files. Standard methodologies and tools exist to deal with the problems of representing the informational contents of a large organization. 4 The first step is the construction of a conceptual schema or model; this is intended to capture the meaning of the information