There are many sources of on-line information available to those responsible for making decisions in complex situations. However, most of that information is either intended for human use or is available only in custom or proprietary formats. Both of these conditions reduce the ability of computer software to perform automatic reasoning about this information. As a result, the usefulness of on-line information in intelligent decision-support systems is currently limited to the few sources that are implemented specifically for those systems. The vast mass of sources that do not fit that description are virtually invisible.

Current practices for integrating multiple information sources have proven too costly to implement and have produced inflexible systems. These practices, which are discussed further below, cannot cope with the magnitude or the diversity of the available information.

This paper describes the TEGRID project, which demonstrates the use of a service oriented architecture to enable a more flexible, loosely coupled system of interoperable information sources and consumers.


Software Engineering



URL: http://digitalcommons.calpoly.edu/cadrc/55