The Open Agent Architecture OAA, developed and used for several years at SRI International, makes it possible for software services to be provided through the cooperative e orts of distributed collections of autonomous agents. Communication and cooperation between agents are brokered by one or more facilitators, which are responsible for matching requests, from users and agents, with descriptions of the capabilities of other agents. Thus, it is not generally required that a user or agent know the identities, locations, or number of other agents involved in satisfying a request. OAA is structured so as to minimize the e ort involved in creating new agents and wrapping" legacy applications, written in various languages and operating on various platforms; to encourage the reuse of existing agents; and to allow for dynamism and exibility i n the makeup of agent communities. Distinguishing features of OAA as compared with related work include extreme exibility in using facilitator-based delegation of complex goals, triggers, and data management requests; agent-based provision of multimodal user interfaces; and built-in support for including the user as a privileged member of the agent community. This paper explains the structure and elements of agent-based systems constructed using OAA. The characteristics and use of each major component o f O AA infrastructure are described, including the agent library, the Interagent Communication Language, capabilities declarations, service requests, facilitation, management of data repositories, and autonomous monitoring using triggers. To provide technical context, we describe the motivations for OAA's design, and situate its features within the realm of alternative software paradigms. A summary is given of OAA-based systems built to date, and brief descriptions are given of several of these.2

The design and development of the Open Agent Architecture (OAA)l system has focused on providing access to agentbased applications through an intelligent, cooperative, distributed, and multimodal agent-based user interfaces. The current multimodal interface supports a mix of spoken language, handwriting and gesture, and is adaptable to the user's preferences, resources and environment.Only the primary user interface agents need run on the local computer, thereby simplifying the task of using a range of applications from a variety of platforms, especiall y low-powered computers such as Personal Digital Assistants (PDAs). An important consideration in the design of the OAA was to facilitate mix-andmatch: to facilitate the reuse of agents in new and unanticipated applications, and to support rapid prototyping by facilitating the replacement of agents by better versions.The utility of the agents and tools developed as part of this ongoing research project has been demonstrated by their use as infrastructure in unrelated projects.

Abstract. In this paper, we discuss how multiple input modalities may be combined to produce more natural user interfaces. To illustrate this technique, we present a prototype map-based application for a travel planning domain. The application is distinguished by a synergistic combination of handwriting, gesture and speech modalities; access to existing data sources including the World Wide Web; and a mobile handheld interface. To implement the described application, a hierarchical distributed network of heterogeneous software agents was augmented by appropriate functionality for developing synergistic multimodal applications.

No abstract