In a context of short-term cooperation, enterprises with complementary skills are dynamically interconnected according to their needs. To deal with requirements in a such cooperation, we present in this paper CoopFlow that aims at providing a useful artifact for preservation of the privacy of workflows partners, pre-established workflows and pre-established workflow management systems. CoopFlow consists of three steps: workflow abstraction and advertisement, workflow matching and interconnection, and workflow cooperation. To preserve the privacy of partners, the first step of the approach includes an abstraction procedure allowing a partial visibility of the partners' workflows. The second step consists of interconnecting existing workflows of partners with complementary skills attributing to a matching procedure that allows checking both behaviors and business semantics of partners' workflows. Finally, the last step of CoopFlow consists of controlling the inter-enterprise workflows cooperation allowing cooperation partners to integrate the existing workflows and to check whenever they can cooperate and change their partners, which frequently leads to support spontaneous, dynamic and short-term cooperation. With this intention, we developed a platform that manages inter-operability by integrating pre-established workflow management systems. The proof of concept is provided by integrating three heterogeneous and open workflow management systems to the CoopFlow platform.

Web services provide standard means to promote interoperability and extensibility among software applications, as well as to allow them to be combined in order to perform more complex operations. Therefore web services have to be described, published, discovered, invoked and composed. Within this process describing Web services using ontology will enhance service discovery. The "quality" of that discovery depends on the used algorithms and techniques. In this paper we propose a novel approach to enhance web service discovery based on, among others, requesters' preferences. First of all, we propose a way to express preferences of request elements. Then we introduce a matching algorithm that takes into account theses preferences in the discovery process.