XML on MOM technology: a new approach for library software

Abstract

In the world of libraries the advent of the Web has produced a revolution that shows no signs of coming to an end. Access to OPACs in the Telnet mode is by now far away and nowadays all library software include a Web OPAC in their package of functions. The Web OPACs however display bibliographic data in a non structured way and this is due to the fact that the structure of the HTML language is rigid and cannot adapt itself to the detail of a bibliographic description. However there are metalanguages with which it is possible to structure documents in a detailed manner.

The progenitor of these metalanguages is SGML (Standard Generalized Mark-up Language) which was born at the end of the Sixties. More recently the W3 Consortium has developed a new metalanguage, XML (eXtensible Markup Language) which, although maintaining the main capacities of SGML, has a more simple syntax. By using XML as a language for defining new types of documents, specialized in bibliographic information, we can reproduce inside them the structure of a database suitable for preserving bibliographic fields (for example, MARC format fields) and, lastly, use these documents as a single platform for the exchange of data between the service provider and the user.

It is no wonder that the Library of Congress has begun a project called MARC/SGML and similar projects for XML are described in various Web sites. Although these metalanguages are extensively studied in the library world, they are by no means widespread in the installed software. The reason for this must however be sought in motivations that are more of a commercial nature than of a technical one; the development of applications able to read XML documents form part of the programmes of the most advanced software producing companies. The advantages of XML in applications for the Web are many:
- more effective search capabilities;
- development of more flexible applications;
- integration of data coming from different sources, through MOM (Message-Oriented Middleware) technology;
- local data processing and manipulation;
- different views of the same data structure;
- intelligent updates;
- greater scalability of the servers.

A simple MOM application, with a three-tier architecture, has been developed at the CBD (Library and Documentation Centre) of the University of Catania on a sample of bibliographic records containing a subset of USMARC fields. The result is similar to a classic OPAC, with the basic difference that the XML file of results carries the MARC structure of the retrieved records.

The future evolution of the application presented here foresees a complete USMARC/XML description and the development of the modules for interaction with the data base even in the writing mode. The goal of this research is not the development of a library automation software package, but the test of a methodology which could prove useful in local applications developed by individual libraries.

Another field where structured languages will find ample application is that of full text documents, with a standard structure, and an experiment is also under way in the CBD for the editing and storing of university dissertations in XML format.