Service-Oriented Architecture (SOA) has emerged as a helpful paradigm for designing, building and using distributed software systems. SOA facilitates establishment of ICT support for business processes, which is readily available, flexible and easily maintainable across multiple organizations and platforms. The concept of service adopted by SOA evolved from modular object/component middleware approaches, such as CORBA, DCOM and J2EE. However, Web services have become the technology of choice for implementing service-oriented software systems, primarily because they are based on ubiquitous Internet standards, such as HTTP and XML, and because they provide a practical solution for ‘loose coupling’. Whereas the uptake of Web services based SOA is impressive, there are still important fundamental challenges not addressed by this technology.
Firstly, the ‘plug and play’ interoperability of Web services to enable ad hoc cooperation of new partners is very limited. For on-demand composition of services in an open service-oriented world, interoperability has to be ensured at different levels (syntactic, semantic and pragmatic) and in different dimensions (information and behavior). Thus, a service request must be processed based on these interoperability requirements to achieve discovery and selection of services, and the selected services have to be composed, adapted and/or mediated enforcing correct interoperability to achieve delivery of the requested service. Current research in this direction is using, for example, Semantic Web and ontology technologies.
Secondly, the property of ‘loose coupling’ is not appropriate for many applications that involve stateful components. Hence, the benefits of Web services and SOA would be very limited for the developers of such applications if they themselves have to solve the issues of stateful interaction, notification of state changes, support for sharing and coordination etc. If we want to relieve the developers, by placing these concerns at the service infrastructure level, what are then the (extended) interoperability requirements and what interworking and coordination architectures would be appropriate? These are questions currently tackled with, amongst others, Grid services technologies that extend Web services with resource sharing and event notification.
Finally, Web services alone are not sufficient to guarantee a proper ‘alignment’ between business requirements and software functionality. What is needed is a structured approach for the development of service-oriented software solutions, in which consistency with business requirements, (de-)composition of application services, and mapping onto (alternative) technology platforms can be systematically and separately addressed. Such an approach, which can be referred to as Service-Oriented Development (SOD), acknowledges that SOA is more than infrastructure; it also forces to consider design and language concerns, and how these are supported with tools. A relevant technology in this area is BPEL4WS.
The goal of this workshop is to focus on the fundamental challenges related to SOA, to discuss what architectural/theoretical foundation is needed, and how this foundation can be supported by new or (extensions of) available technologies and infrastructures. The workshop aims at contributing to the dissemination of research results, establishment of a better understanding, and identification of new challenges, by bringing together academic and industrial researchers interested in the future of service-oriented architecture.