In the last few years, information retrieval has expanded beyond its traditional home in libraries and formal information systems. It is now firmly in the hands of end-users, and is still undergoing a searching revolution which started in the late 1980s, when growth in the publication rate of CD_ROM databases began making the old models of information retrieval obsolescent.
Then, all database producers developed unique and increasingly complex command systems for information retrieval, and users were normally professional searchers, trained in the use of whatever retrieval interfaces were provided. That old model suited a seller's market, where:
- computerised databases were rare and expensive providers of quality information
- competition between databases was rare
- user's time and training resources were freely available, and
- searchers had little choice but to make the best possible use of the few accessible databases.
In today's buyers' market for information retrieval, electronic information is normally cheap and readily available but:
- its quality may be unreliable
- databases are numerous but customers are rarely trained to use them, and
- these customers must often make their choice from among many competing information retrieval systems
The old database producers competed by acquiring more files of data and adding sophisticated, but rarely used, search options to their unique search interfaces. Thus, searchers had to make their behaviour fit the interfaces. Now information systems must fit themselves to the universal WWW browser interface, and customers quickly move from one system for another if the first doesn't deliver the right results. In this model of information retrieval, the end-user is customer and king. Database producers no longer have captive audiences tied to their products by the cost of their experience and learning. In addition, they must attract customers from rival databases who do not want to spend time learning to use new interfaces. Today's information retrieval systems usually have interfaces that are so simple and so similar that new customers can rapidly teach themselves how to use most of the system. As such, an arsenal of sophisticated search options is no longer required.
The aims of this module are to review the results of these changes, to critically examine the current state of the Internet'r major information retrieval channels, and to analyse the relationship between Web page features and their information retrieval consequences.