Correspondence and Translation for Heterogeneous Data

A primary motivation for new database technology is to provide support for the broad spectrum of multimedia data available notably through the network. These data are stored under different formats: SQL or ODMG (in databases), SGML or LaTex (documents), DX formats (scientific data), Step (CAD/CAM data), etc. In this paper, we provide a formal foundation to facilitate the integration of such heterogeneous data and the maintenance of heterogeneous replicated data. We introduce a middleware data model. The data structure is based on ordered labeled trees. It is general enough to capture the essence of formats we are interested in. It is simple enough so that correspondences/relationships within existing data can be described declaratively using simple correspondence rules. The rules may use standard tree constructors (cons and merge), and tree rewriting takes into consideration essential properties of data collections. We also consider the problem of data translation from one world to the..