Semantic Web Services Languages and Technologies: Comparison and Discussion

Web services were primarily designed to be loosely coupled and to provide inter-operability between business applications. The large amount of human interaction, that is required for integration of multiple applications by the current Web service technologies, needs to be reduced. To make integration automated we need machine processable description of the data and functions of the involved entities. Semantic Web technologies use description logic based languages to annotate the content with ontology concepts to make it machine understandable. Due to the increasing interest in knowledge than data, and the rising popularity of the Semantic Web and Web Services, there have been significant interests in developing technologies that support Semantic Web Services. The Semantic Web industry is experiencing a need for identifying and developing technology that will provide a firm and long-term foundation for the future of Web services on the internet. This foundation is required to support the universal approaches to service deployment and use. It should support the currently technically feasible approaches; and should have the features of flexibility, extensibility, and consistency with the vision of the Semantic Web. A variety of tools and modeling frameworks that support publication, discovery and composition of Semantic Web Services have been developed in the near past. These initiatives include OWL-S [Martin et al., 2003] (formerly DAML-S [DAML-S]), METEOR-S [METEOR-S, 2004], and WSMO [Roman et al., 2004] etc. In general however we feel that no tool/framework provides all that is required for a generic Web services modeling platform for the Semantic Web. These standards are still incomplete and may not fulfill the future demands of the industry like complexity, scalability, dependability, to name a few. The SWSI Language Committee [SWSL] is working on identifying and developing a computer language technology which will standardize the semantic information about web services and develop a language for its declarative specification. Also, the semantic information about web services needs to be general enough to be able to support nontrivial mechanized interactions between Web services and intelligent software agents. Ideally, the Semantic Web services language should enable: • dynamism in all aspects of Web service use, such as selection, discovery, composition, invocation, negotiation, and recovery from failure; • be extensible and should be closely integrated with knowledge resources on the Semantic Web.