Distributed Objects with Sense of Direction

An object system consists of a collection of objects and their relations; each object has a state (e.g., local variables) and a behavior (set of actions it may execute) and its global behavior is described in terms of interactions between its objects. When dealing with object systems, the manipulation of names is crucial to such an extent that an objects has been defined as “any entity in a system which deserves a name ...” [8]. Such a definition has been used in the context of distributed operating systems but is clearly applicable to any other domain. Furthermore, some authors have stated that the discipline of object-oriented programming is all about manipulating object names (e.g., [9]). A naming scheme is a mechanism used for naming objects and manipulating objects through their names. In object systems, some of the main objectives of a naming scheme are: to designate an object, to distinguish between objects independently of their state, to communicate to other objects the knowledge about some objects. Furthermore, a naming scheme should be stable and reliable, that is, names must remain valid and keep denoting the same objects when the system changes its state. In order to achieve these goals, the existing naming schemes either use an approach based on global names (e.g., each object uses as its name an intrinsic property which distinguishes it from the other objects) or they use hierarchical names based on the location of the objects (e.g., email addresses). On the other hand, locality of names and independence on the location are two very desirable characteristics of a naming scheme, and to date no proposal is available with such properties. This work has been partially supported by NSERC and FCAR