 Download PDF version |
| Scope and Objectives |
|
Web Services are today the default choice for solving the various issues related to the organization and implementation of software distribution. The attention on this new paradigm is mainly based (a) on the benefits in terms of system interoperability stemming from WSs and (b) on intra- and inter-organisational accounting reasons. Technical solutions are becoming numerous and in the next future we can figure out that the further adoption of common agreed specifications, involving the diverse phases of WS development and description, will raise even more the interest from academia and industries. However, while many current systems are concerned with the migration towards the WS technology, well defined methodologies for modeling service-oriented applications, and the impact of this paradigm for what concerns analysis capability, remain still largely unclear. Indeed, analysis activities are obviously strongly influenced by the type and amount of information that is attached to the description of a WS and consequently by the tools that can be used. Modeling WS based systems seems particularly challenging considering that most of the web services that will constitute the final system are implemented and hosted by different stakeholders and also evolve independently. The modeling of interactions needs information coming from the documentation of involved services, but the current approaches are still not adequately equipped to accomplish this task. The current methodology of specifying WSs is generally recognized as poor and various proposals to augment the descriptions of single and choreographed services are on their way. In this process we claim that considerations concerning testability and the capability to base them on an increased information model should have a first-class role. In fact, testing of WS-based applications is hindered by different factors, such as the inherent scarce possibility of controlling the different pieces of an application (that could reside in hosts owned by different organizations) and the high dynamism, that permits to retrieve and bind those pieces at run time. Therefore, the development of reliable WS-oriented systems will require suitable mechanisms, tools, and techniques, also based on appropriate ways of describing the services that may not yet have been taken into account, even by those consortia and standards bodies who are already paving the way to identify common notations for WS specification. An interesting aspect of WS technologies is the emergence of a standardization effort within different application community to guarantee correct interactions. Such effort generally aims at precisely describing the interface and the expected behaviour of a service. Probably the most remarkable example today is that of Grid Computing, but similar effort can be noticed within other communities such as for instance that of e-Learning, in which standard service oriented architectures are becoming to be made available by the IMS global consortium. The definition of standard services requires the establishment of an agreed process/procedure for modeling and documenting WS-based systems. In fact such information become of vital importance for WS developers that want to implement their version of the standards, and to certification agencies that will assess the conformance to such standards with several methods; the most relevant certainly being testing. This workshop aims at putting together researchers and practitioners to discuss the different issues related with modelisation and testing of Web Services. Particularly interesting will be examples of processes followed in the implementation of real service based systems. Some more precise topics of interest for the workshop, but not intended as exclusive, are:
|