The explosive growth of the Web requires servers to be extensible and configurable. This paper describes our adaptive web server, OpenWebServer that employs a metaarchitecture. It supports dynamic adaptation of feasible design decisions in the web server design space by specifying and coordinating metaobjects that represent various aspects within the web server. We present some examples of system adaptation that change and tune configuration of concurrency, caching, logging, load balancing and fault tolerance. OpenWebServer can evolve continually beyond static and monolithic servers.

Unified Modeling Language (UML) has been widely accepted in the software engineering area, because it provides most of the concepts and notations that are essential for documenting object-oriented models. However, UML does not have an explicit format to describe and interchange its model information intentionally. This paper addresses the UML model interchange and presents our efforts to make UML highly interoperable.

A number of approaches have been proposed to provide support for crosscutting concerns at the design level. This paper compares some of these design approaches. A set of “good design ” criteria is used to identify strengths and weaknesses of the individual approaches. The paper concludes with a proposition to unify the strengths of the discussed approaches into a more effective model. 1

Architectural descriptions of component style and behavior provide powerful predictive capabilities for interoperability conflicts. In this paper, we describe architectural indicators for a first pass interoperability assessment. These indicators form a set that is minimal and realizable, thus having the potential for automated forecasting. We encapsulate these indicators into a XML architecture interaction conspectus (XMLAIC) that fronts each independent software system. XML provides an open, standards-based way to implement the conspectus, affording industry-wide interoperability assessment for both pre- and post-component purchase.

The emergence of the Internet has radically changed the way in which we learn, teach and train. This paper proposes an integrated and extensible architecture for agent-based distance learning. The architecture provides component-based extensibility, allowing emerging technologies to be plugged-in, so that they can produce synergy. It provides HTTP and IIOP connections for maintaining and delivering courseware to students. Via the HTTP connection, our Persona system provides the personalization service to each set of courseware, allowing it to customize its content and/or presentation contextsensitively. This service facilitates the effective delivery of courseware. Via the IIOP connection, our SoftDock system provides the foundation facility allowing users to work collaboratively in teams. Our current educational domain is software modeling. Participants learn the basic concepts and principles of software modeling, and then leverage their newly-acquired modeling skills. We believe our work provides a blue print for showing how emerging technologies can be applied to practical distance learning applications.