Introduction to the Special Section: The Role of Semantic Web Services in Enterprise Application Integration and E-Commerce

Christoph Bussler, Dieter Fensel, and Norman M. Sadeh, Guest Editors
International Journal of Electronic Commerce,
Volume 9, Number 2, Winter 2004-05, pp. 7.

This Special Section of the International Journal of Electronic Commerce focuses on the proposed intersection of three domains that have very recently begun to draw enormous attention throughout academia and industry and are of the utmost importance as well as relevance for computer science and the business world:

The promise is that the combination of Web service technology with Semantic Web technology (“Semantic Web services”) will make enterprise integration dynamically possible for enterprises of all types and sizes, unlike the “traditional” technologies, such as electronic data interchange (EDI) or value-added networks (VANs), that are considered static and difficult to manage. In addition, enterprise integration will become more reliable and easier to achieve without the tedious low-level implementation problems that can be observed in today’s approaches.

There are a few current efforts to define a coherent model of Semantic Web services. One is OWL-S as part of the U.S.-based DAML program [1]. Another, more recent approach is WSMO (Web Service Modeling Ontology), an open European effort initiated by the Digital Enterprise Research Institute (DERI) [2, 3]. These efforts describe Web services semantically using Semantic Web languages. Both efforts, for example, introduce the notions of “process” and “Web Service composition” to describe behavior. OWL-S has a process model. WSMO, in contrast, distinguishes orchestration (problem-solving through Web service invocation) from choreography (communication pattern through message exchanges). However, there are differences, too. For example, WSMO proposes the explicit representation of goals and capabilities for dynamic discovery. Also, data and process mediation are explicit concepts applied by Semantic Web services to overcome the heterogeneity problem in the real Web world.

What has not yet been addressed by OWL-S and WSMO is the development of a methodology for designing an actual integration scenario like travel planning or goods purchasing. A software engineering methodology that defines the design steps to be taken in order to derive a working Semantic Web service integration is out of the scope of and is yet to be addressed by OWL-S and WSMO.

The articles in the Special Section address some of the areas of Semantic Web services.

“Harmonise: A Step Toward an Interoperable E-Tourism Marketplace” by Oliver Fodor and Hannes Werthner analyzes and discusses data interoperability and data mediation in the context of the tourism industry as an application domain. The many information systems used in this industry contribute to a large number of different ontologies that are all different conceptualizations of travel industry concepts. Because all the parties interoperate, a solution that enables data interoperability is essential for this industry. The article shows how the data interoperability problem in the tourism industry can be solved using Semantic Web technologies, mainly ontology languages, by an indirect approach. Instead of mapping the various ontologies directly to the involved parties, a general tourism industry ontology is constructed. Every party maps to this general ontology. This allows mapping from any party’s ontology to any other’s party ontology indirectly through the general tourism ontology.

“Systematic Approaches for Designing B2B Applications” by Tariq Al-Naeem, Fethi Rabhi, Boualem Benatallah, and Pradeep Ray introduces a design methodology for an organization that has to define and develop its business-to-business (B2B) connectivity with other organizations. A running example of interenterprise e-payment is introduced that deals with service providers, service consumers, and their respective banks and interbank clearing organizations. There are several elements that have to be decided. These include alternative process models that define the interactions between trading partners (e.g., direct or indirect communication) and architectural models that define the type of communication and technology choices to be made when selecting implementation technologies like programming languages and communication mechanisms. These elements are discussed in great detail to show the complexity that has to be dealt with. Based on the possible choices of process models, architecture models, and implementation technologies, a design process methodology is presented. This methodology, working in terms of specific B2B integration needs, outlines the steps that can be taken to derive a successful B2B integration with trading partners.

In “Framework for Semantic Web Process Composition,” Kaarthik Sivashanmugam, John Miller, Amit Sheth, and Kunal Verma describe the concept of semantic process templates. These constitute a workflow-style approach to modeling semantic processes. Activities, data flow, and control flow are the prominent concepts in this approach. In addition, Web services can be invoked from semantic process templates, as is illustrated by the example of product purchasing that is discussed throughout the article. Web services have to be selected in order to be invoked. A discovery mechanism based on semantic matching and quality of service matching is provided for this purpose. At design time, a designer can request a list of all matching Web services from repositories and select the appropriate one to be put into the semantic process template. Finally, once a Semantic Web template has been established, executable processes can be generated. The approach currently supports generation into a Business Process Execution Language (BPEL) environment, but the authors state that their own process-execution implementation is well under way.

“JOpera: A Toolkit for Efficient Visual Composition of Web Services” by Cesare Pautasso and Gustavo Alonso explores aspects of the need for visual tools in the modeling of processes. A visual process-composition language is introduced that consists of symbols for workflow steps and subworkflows, data flows, and control flows. The functional capability of this visual language, a mapping to BPEL and from BPEL, allows the graphical modeling of processes and their translation to a BPEL execution environment. JOpera, however, provides its own execution environment, too.

Finally, deployment alternatives are discussed that make it possible to place various components on different computing nodes and to select different storage mechanisms. Detailed measurement results are shown to demonstrate the performance of the JOpera execution environment.

REFERENCES

  1. www.daml.org/services/.
  2. www.wsmo.org.
  3. www.deri.org.

CHRISTOPH BUSSLER (Chris.Bussler@deri.org) is Science Foundation Ireland Professor at the National University of Ireland, Galway, executive director of the Digital Enterprise Research Institute, and leader of its Semantic Web services research group. Before taking this position, as a member of Oracle’s Integration Platform Architecture Group based in Redwood Shores, California, he was responsible for the architecture of Oracle’s next-generation integration product providing EAI, B2B, and ASP integration. Before joining Oracle, he was at Jamcracker, Cupertino, responsible for defining its ASP aggregation architecture; Netfish Technologies (acquired by IONA), Santa Clara, responsible for its B2B integration server; the Boeing Co., Seattle, leading Boeing’s workflow research; and Digital Equipment Corporation (acquired by Compaq, acquired by Hewlett-Packard), Mountain View, defining the policy-resolution component of Digital’s workflow product. Professor Bussler has a Ph.D. in computer science from the University of Erlangen, Germany, and a master’s degree in computer science from the Technical University of Munich, Germany. He is the author of B2B Integration, two books on workflow management, and more than 60 research papers in journals and academic conferences. He has given tutorials on several topics, including B2B integration and workflow management, and has been the keynote speaker at many conferences and workshops.

DIETER FENSEL (Dieter.Fensel@deri.org) has a diploma in social science from the Free University of Berlin and a diploma in computer science from the Technical University of Berlin (both 1989). In 1993, he was awarded a doctorate in economic science (Dr. rer. pol.) at the University of Karlsruhe, and in 1998, he received his habilitation in applied computer science. In 2002, after serving on the faculties of the University of Karlsruhe, the University of Amsterdam, and the Vrije Universiteit Amsterdam, he took a chair at the University of Innsbruck, Austria. In 2003, he became the scientific director of the Digital Enterprise Research Institute at the National University of Ireland, Galway, based on a large grant acquired from Science Foundation Ireland (SFI). His current research interests include ontologies, Semantic Web, Web services, knowledge management, enterprise application integration, and electronic commerce. He has published around 150 papers as journal, book, conference, and workshop contributions. He has co-organized around 150 scientific workshops and conferences and has edited several special issues of scientific journals. He is an associate editor of Knowledge and Information Systems: An International Journal (KAIS), IEEE Intelligent Systems, Electronic Transactions on Artificial Intelligence, Web Intelligence and Agent Systems, Elsevier’s Journal of Web Semantics: Science, Services and Agents on the World Wide Web, and the Lecture Notes in Computer Science subline entitled “Semantics in Data Management.” He has been involved in several national and international research projects and gave more than 50 courses and tutorials at Berufsakademien, Fachhochschulen, universities, and conferences. Dieter Fensel is the coauthor of Intelligent Information Integration in B2B Electronic Commerce (2002), Ontologies: Silver Bullet for Knowledge Management and Electronic Commerce (2001), Problem-Solving Methods: Understanding, Development, Description, and Reuse (2000), and The Knowledge Acquisition and Representation Language KARL (1995).

NORMAN M. SADEH (sadeh@cs.cmu.edu) is an associate professor in the School of Computer Science at Carnegie Mellon University. He is director of CMU’s Mobile Commerce Laboratory and e-Supply Chain Management Laboratory, and codirector of its newly established Privacy Technologies Center and its Ph.D. program in computation, organizations, and society. He currently teaches, consults, and conducts research on mobile commerce, pervasive computing, privacy, agent technologies, the Semantic Web, and supply-chain management, and more generally is interested in the broader business, social, and policy implications associated with the emerging information society. In the late 1990s, he was program manager at the European Commission in Brussels, where he last served as chief scientist of the Eur0550M European research initiative in “New Methods of Work and eCommerce.” Dr. Sadeh has been on the faculty at CMU since 1991. He has a Ph.D. in computer science from Carnegie Mellon University with a minor from the Business School. He has received numerous scientific and technical awards, has authored well over 100 scientific publications, and serves on the editorial board of several journals, including Autonomous Agents and Multi-agent Systems and Electronic Commerce Research Applications. He is the lead designer of the international Supply Chain Trading Agent Competition and an APICS certified fellow.