Accessing Databases Through the World Wide Web: Issues and Current Practice

This paper examines the issues, current practices and trends for the future regarding accessing traditional databases through the World Wide Web (WWW). It is often easy to forget that the WWW has no data management facilities of its own; it is simply a “pointing” mechanism. Databases have historically focused on rich toolsets for describing, updating, and modelling data. WWW adds to this a platform independent, client-server access method that uses ubiquitous transport medium. The incorporation of these technologies has introduced a number of new challenges, and resurrected some older computing issues. However, the eventual result will be greater information access to a much wider user community. 1.0 Introduction The World Wide Web (WWW) is a popular client-server information system architecture built on a number of Internet Protocols [1]. Although work on WWW began in the late 1980s, it was not until the arrival of the client program “NCSA Mosaic” in 1993 that interest in WWW really took off [2]. WWW servers have increased from 130 in June, 1993 to over 23,000 measured in June, 1995 [3]. In contrast, database management systems (DBMS) appeared in the mid 1960s with the most popular language for DBMSs, Structured Query Language (SQL), originally defined in the mid 1970s [4]. Given nearly 30 years of development and applicability to numerous disciplines, it is not surprising that database technology has achieved its current prevalence. Since these two technology areas, WWW and DBMS, are applicable to such wide audiences, it is not surprising that work is well underway on gatewaying between them. Each technology has strengths and weaknesses of its own, and the combination of the two introduces a new set of strengths and weaknesses. 2.0 Computer System Trends This section examines a brief history of each technology, including the current trends that are leading up to the present synthesis. 2.1 Trends in Database Management Systems