Friday, 3:30–5:00 PM
Chair: Dennis Fetterly
Redundancy Detection in Service-Oriented Systems
Peep Küngas, Marlon Dumas
We investigate the problem of detecting data redundancy from interface descriptions in large service-oriented information systems. By redundant data, we mean data maintained at different locations independently. Such data redundancy may lead to inconsistency and may hinder the reuse of services. We define two simple but effective heuristics that help to identify and pinpoint potential data redundancy. The proposed heuristics have been tested on a federated governmental information system consisting of about 60 independently-maintained information systems providing altogether about 1000 data services described by means of WSDL. The quality of the heuristics is evaluated in terms of precision and recall.
Protocol-Aware Matching of Web Service Interfaces for Adapter Development
Hamid Motahari, GuangYuan Xu, Boualem Benatallah
With the rapid growth in the number of online Web services, the problem of service adaptation has received significant attention. In matching and adaptation, the functional description of services including interface and data as well as behavioral descriptions are important. Existing work on matching and adaptation focuses either on service interfaces or behavioral descriptions. In this paper, we present a semi-automated matching approach that considers both service descriptions: we introduce two protocol-aware service interface matching algorithms, i.e. depth-based interface matching and iterative reference-based interface matching. These algorithms refine the results of interface matching by incorporating the ordering constraints imposed by business protocol definitions on service operations. We have implemented a prototype and performed experiments using the specification of synthetic and real-world Web services. Experiments show that the proposed approaches lead to a significant improvement in the quality of matching between services.
Volunteer Computing: A Model of the Factors Determining Contribution to Community-based Scientific Research
Authors: Oded Nov, David Anderson, Ofer Arazy
Volunteer computing is a powerful way to harness distributed resources to perform large-scale tasks, similarly to other types of community-based initiatives. Volunteer computing is based on two pillars: the first is computational – allocating and managing large computing tasks; the second is participative—making large numbers of individuals volunteer their computer resources to a project. While the computational aspects of volunteer computing received much research attention, the participative aspect remains largely unexplored. In this study we aim to address this gap: by drawing on social psychology and online communities research, we develop and test a three-dimensional model of the factors determining volunteer computing users’ contribution. We investigate one of the largest volunteer computing projects—SETI@home—by linking survey data about contributors’ motivations to their activity logs. Our findings highlight the differences between volunteer computing and other forms of community-based projects, and reveal the intricate relationship between individual motivations, social affiliation, tenure in the project, and resource contribution. Implications for research and practice are discussed.