Technology and Means of Access to Distributed Resources of Databases and Digital Libraries

O.L. Golitsyna, N.V. Maximov
Russian State Humanitarian University

If the digital library is viewed as a complex of inter-related information resources and means of access to them, we can say that the information components of such system form a three level structure, including:

• On metadata level – reference and authority files (heading lists, thesauruses, author, subject and other indices), catalogues and databases;
• On secondary information level – electronic catalogues or bibliographic databases;
• On primary information level – full text information resources;
• In terms of localisation and accessibility there can be two types of resources:
• Localised resources (metadata, content and search facilities combined in the LAN);
• Distributed resources (the above components are physically located in different places).

This paper looks at the information technology aspects of information search in a complex of inter-related resources in the context of digital library created by INION.

The information resource access management is organised via user catalogue containing the information on authorisation levels assigned to different users. The access control server is a specialised system with the following functional databases:

• catalogues of information resources
• catalogue of users
• technological database of current transactions
• statistical database.

The information components of the library include:

• Reference databases
• Bibliographic databases
• Full text resources.

Each of the above components is implemented as a documentary database. The system can generate dynamic links with user defined parameters.

The main role of the system is IT support of the user on different stages of information search, including:

1. Definition of the search topic;
2. Search request formation;
3. Document selection by pre-defined criteria;
4. Search result management;
5. Search results assessment;
6. Search refining.

There are certain difficulties in user interaction with search systems for the latter in many ways exist in old interactive form where the mediation of an information specialist was required. Therefore the users tend to stick to either of the two extremes: complete mistrust of automatic search systems as utterly useless, or complete reliance on them as universal intelligent mechanisms. Neither of the two is correct – search systems are only tools used by people in search for information, but not intelligent automatic search engines.

All the systems of automatic information search fall in two major groups: classic type systems originally developed on the basis of library systems and hypertext systems. The two types differ in scenarios and tools of relevant information selection by the user. The systems of the first type take little notice of the context and the user has to sift the search results very carefully. Systems of the second type are much more context sensitive.