Designing Intelligent Adaptive Distributed Systems is an open research issue addressing nowadays technologies such as Communicating Objects (COT) and the Internet of Things (IoT) that increasingly contribute to our daily life (mobile phones, computers, home automation, etc.). Complexity and sophistication of those systems make them hard to understand and to master by human users, in particular end-users and developers. Those are very often involved in learning processes that capture all their attention while being of little interest for them. To alleviate human interaction with such systems and help developers to produce them, we propose WOF, an object oriented framework founded on the concept of \emph{Wise Object} (WO). A WO is a software-based entity that is able to learn on itself and also on the others (e.g. its environment). Wisdom refers to the experience (on its own behavior and on the usage done of it) such object acquires by its own during its life. In the paper, we present the WOF conceptual architecture and the Java implementation we built from it. Requirements and design principles of wise systems are presented. To provide application (e.g. home automation system) developers with relevant support, we designed WOF with the minimum intrusion in the application source code. The adaptiveness, intelligence and distribution related mechanisms defined in WOF are inherited by application classes. In our Java implementation of WOF, object classes produced by a developer inherit the behavior of Wise Object (WO) class. An instantiated system is then a Wise Object System (WOS) composed of wise objects that interact through an event bus according to publish-subscribe design pattern.