Abstract not found

Mobile code is being investigated by many researchers as an alternative for the implementation of more flexible distributed applications. A number of programming paradigms based on mobile code have been developed. This paper focuses on implementation of distributed services using the messenger paradigm. With the messenger approach which combines the potential of mobile code, runtime code generation and code interpretation, it is no longer necessary to know the service interface at development time in order to implement a client that can interact appropriately with the service. The service is published with its operational interface. The client interprets, at run time, the service's operational interface and collects from it the information required to interact appropriately with the service. Keywords: distributed service, messenger,messenger interface description language. INTRODUCTION Distributed applications are expected to run in large, heterogeneous, asynchronous and dynamic envi...

. 1 Mobile agents are usually expected to execute in open environments. This openness implies that they should be able to dynamically learn how to interact with other agents and services which were not known at development time. The interlocutors therefore have to publish enough information about their functionality, while at the same time they have to restrict access rights in order to preserve their integrity. We describe in this paper a messenger-based framework which proposes run-time generated interfaces to address this duality. 1

this paper is to describe novel applications of Mobile Code technology which have not appeared yet but should be feasible with our current knowledge of the domain. These new applications contradict the often-made observation that Mobile Code is just another technique that does not really bring much more possibilities than existing technologies for distributed applications. There is a whole class of problems that have not received much attention yet and that are not well managed by current environments. These are the problems of maintaining consistency of dynamic information and maintaining systems in adequacy with the ever changing requirements of customers. Our motivation is that, besides the quantitative improvements that most people expect from using Mobile Code, there is also a qualitative benefit which is even more important but not universally recognized now: Mobile Code allows communication with less conventions than message passing [5, 3]. Processes interconnected by Mobile Code still have to agree on high level encoding and synchronization primitives but these agreements are only a fraction of what is necessary to communicate. Many context dependent aspects can be encapsulated inside Mobile Code and changed when the context changes. Encapsulation has the same benefits here as in other software engineering domains: it reduces the dependency between components, thus reducing the number of modifications that we must make to software in order to adapt it to new requirements. For this reason we think that it is the best way to cope with systems that are distributed, hence not manageable by a single person or organization; that are dynamic, because the information they contain must change when the world itself changes; and that are evolving since the users discover n...