For many years I have worked on developing relational models for museum information based first on notions developed in the work of the Common Agenda for History Museums and then on a much simplified version of the ICOM-CIDOC model long under consideration by the Reconciliation of Standards Working Group of the Documentation Committee of ICOM. A draft of that simple model was developed at Paris two years ago, and presented at the last ICOM meeting, but never approved by the committee as a whole and so not released. Nevertheless, it follows the structure of what you've been ddoing with LASSI and others have done in developing data structures. I have used FoxPro as the base program and for the basic data structures, and I have been pleased with what it can do. It handles text easily with no apparent constraints until you begin trying to imbed a few Greek characters. It works happily in DOS and Windows environments, and now is extended to MAC and eventually UNIX as well. While one may argue whether xbase tables are ever truly relational, in my experience they can be made to behave as if they were. In short, I think that the issues you raise are answerable with simple tools thoughtfully used. Whether these are tools that suit the scale of what you undertake is another matter. Jim Blackaby | [log in to unmask]