The current landscape of Agent Communication Languages

Despite the substantial number of multi-agent systems that use an Agent Communication Language (ACL) the dust has not settled yet over the landscape of ACLs. The semantic specification issues have monopolized the debate at the expense of other important pragmatic issues that must be adequately resolved in the immediate future if ACLs are going to support the development of robust agent systems. After introducing some of the basic concepts relating to Agent Communication Languages, we cover KQML and FIPA ACL, the two existing fully-specified ACLs. We give a brief introduction to their semantics and the issues relating to semantic descriptions of ACLs. We then shift our focus beyond the semantics and point to problems and limitations shared by both ACLs. Questions such as the nature of conformance of an agent system with an ACL specification and issues such as naming, registration, authentication, basic facilitation services, etc., may or may not be (technically speaking) part of an ACL specification, but we feel that the answers and solutions to such problems can “make or break” an ACL. We finally discuss the future of ACL standardization efforts and identify the issues that are likely to emerge as we gain experience in building and deploying agent-based systems. Final draft of a paper to appear in IEEE Intelligent Systems, volume 14, number 2, March/April, 1999.