ADOME, an ADvanced Object Modeling Environment, was developed to better understand, and to more adequately accommodate, the requirements of next-generation information systems (NGISs) and their applications. While ADOME has been very effective in capturing declarative knowledge semantics, it still falls short of capturing active aspects of applications. Such active capabilities are particularly important for such NGIS applications as CSCW where the form of cooperation/collaboration is often realized as activities....[ Read more ]

ADOME, an ADvanced Object Modeling Environment, was developed to better understand, and to more adequately accommodate, the requirements of next-generation information systems (NGISs) and their applications. While ADOME has been very effective in capturing declarative knowledge semantics, it still falls short of capturing active aspects of applications. Such active capabilities are particularly important for such NGIS applications as CSCW where the form of cooperation/collaboration is often realized as activities.

Thus, to extend the scope of ADOME so as to better support those active aspects, reactive capabilities (rules and events) are introduced into it. In this thesis, an extension to the data/knowledge model of ADOME is introduced which incorporates reactive capabilities without sacrificing the flexibility of rule sharing in ADOME. The resultant system is called ADOME-II for obvious reasons. Rules and events are modelled as first-class objects. ADOME-II has a rich set of pre-defined composite event expressions and a well-defined execution model. An approach to reactive processing is also introduced so that the reactive capabilities can be integrated into the loosely-coupled architecture of ADOME. In this approach, production rules combined with methods are used as a unifying mechanism to process rules, to enable composite events to be detected incrementally, and to allow new composite event expressions to be introduced into the system in a declarative manner. Methods for supporting new composite event expressions and handling other parameter contexts are outlined to illustrate the extensibility of the approach. Other relevant approaches are compared and it is concluded that the approach of ADOME-II is more declarative and extensible. Such extensibility is critical for the system to be able to evolve in accommodating dynamic NGIS applications' requirements.