Deploying intelligent systems on a global scale

Intelligent systems are rarely deployed in isolation. Furthermore, with the advent of the global Internet, corporate intranets, and the Web, the potential for interactions between intelligent systems has never been greater. However, new technologies such as Web services, the semantic Web, grid computing, and peer-to-peer networks have made it easier than ever to create intelligent applications that span not only multiple machines but also multiple sites and organizations. So, one great challenge in intelligent systems research is to harness these technologies to verify and apply longstanding theories from distributed AI, agent systems, and other areas of networked intelligent systems research. Although we can generate knowledge from models, simulations, and small-scale trials, gaining a thorough understanding of intelligent systems requires increasingly large-scale deployments in challenging environments. Unfortunately, such experimentation is time consuming and costly-often requiring significant infrastructure development, maintenance, and low-level implementation before we can tackle more challenging and innovative elements. As the environments in which we employ intelligent systems continue to grow, so do the costs of deploying, testing, and verifying such systems.