CLARITY - A NEW ENVIRONMENT FOR GENERALISATION USING AGENTS, JAVA, XML AND TOPOLOGY
Paul Hardy, Melanie Hayles, Patrick Revell · 2003
Over the past few years, many national and regional mapping agencies have built a basic scale dataset covering their territory, often now referred to as ‘framework ’ data. At the same time, there have been increasing requirements for rapid creation of multiple products at different scales and specifications, combined with great pressures to cut costs and reduce staffing. These conflicting requirements have given a real incentive to use automated generalisation. However, production-quality, intelligent tools for contextual generalisation have been scarce or unavailable. The advantages of object data modelling, active object toolkits and software agents have been proven by research projects, and by successful deployments of prototypes in production flowlines at national mapping agencies. Work is now underway, guided by the MAGNET consortium to develop a new production-ready object generalisation environment (Clarity). This is developed from the AGENT prototype, but building on experience gained, and adopting standard technologies such as XML and Java. This paper overviews the current and future developments and product strategy of a leading automated solution for cost-effective object and agent generalisation. It particularly describes the new architecture with its use of XML and Java, and the importance of topology. 1