The following is an example of an operational demonstration scenario that illustrates inter-agent communication (see Figure 2).
The user tells the interface agent (in spoken language) that ``When mail arrives for me about a security break, get it to me". The interface agent translates this statement into a logical expression, and posts the expression to the blackboard. The blackboard server determines that a trigger should be installed on the mail agent, causing it to poll the user's mail database. Once the mail agent has determined that a message matching the requested topic has arrived for the user, it posts a query to find out the user's current location. The calendar agent responds, noting that the user is supposed to be in a meeting which is being held in a particular room; the database agent is then queried for the phone number of the room. Finally, the telephone agent is instructed to call the number, ask for the user (using voice synthesis), perform an identification verification by requesting a touchtone password, and then read the message to the user. We intend to add agents that would increase the number of ways in which a user might be contacted: agents to control fax machines, automatic pagers, and a notify agent that uses planning to determine which communication method is most appropriate in a given situation.