Future Trends

---

Agent-Supported Web Services

Service-oriented computing is becoming the prominent paradigm for distributed computing and is creating opportunities for service providers and application developers to develop value-added services by combining Web services. Web services technology is currently being touted as the ideal solution to meet the requirements for the dynamic composition of Enterprise Information Systems (Yang & Papazoglou, 2000).

Agents have the potential to harmonize Web services’ behaviors. The design of many software agents is based on the assumption that the user needs to specify a high-level goal instead of issue explicit instructions, leaving the how and when decisions to the agent. A software agent exhibits a number of features that make it different from other traditional components (Jennings et al., 1998), including autonomy, goal orientation, collaboration, flexibility, self-starting ability, temporal continuity, character, communication, adaptation, and mobility.

Software agents can play both roles of a Web service client as well as the role of a Web service. As a client of service, an agent can perform searches of different entries stored in a UDDI. It then can make message- and RPC-style calls to a Web service. As a Web service, an agent has a dual nature that combines the characteristics of the two technologies: the ability to be published, found, and called as a Web service and the ability to make autonomous decisions.

An agent-supported Web service can include a local information space and a set of agents supporting the Web service.

A service registry is the medium and services provider for service discovery.

The PA, upon receiving a task from a user, requests a task agent (TA), initiates a message, and posts it onto the nearest service registry. The message contains pertinent information, such as the job task, the requirements and criteria for the job, the expiry date, and ways of contacting the PA.

At the same time, the first respondent (a TA) interested in accepting the offer indicates that the case is closed to prevent other service providers from responding to the same message. Subsequently, the first responding TA then keeps in touch with the concerned PA for finer negotiation. When all the qualified TAs are preoccupied, the message remains outstanding on the service registry until it attracts an interested TA that is free. For certain tasks that require a number of service providers bidding for a specific job, the interested TAs contact the concerned PA, which filters and selects the vendor based on certain criteria, such as the quoted price and the qualifications.

Web services for distributed learning include knowledge management and information-resource management. Knowledge-management Web services manage domain knowledge (ontologies, concepts, etc.) and knowledge about curriculum planning.

Information resource management Web services include learning object- repositories management, instructor-information management, tutor-information management, and student-information management. These information- resource management Web services are responsible for getting information about the resources needed.

To provide efficient and effective Web services, some agents can be deployed to support the Web services, taking advantage of the autonomy and distribution of agents. For example, a “spider-like” broken-link-checking agent can be used to maintain a LOR, supporting the LORs’ management Web service.

Another example is the Ontology Web service. Ontology is a taxonomy database used for a target language for (a) the terms in the prerequisite and postconditions of learning objects, and (b) the terms in the learner profiles. An agent-supported ontology Web service can maintain the ontology knowledge autonomously when needed.

Knowledge Management

Ontology-Based Domain Modeling

There are two aspects to the obstacles of agent technology in distributed learning. One is the difficulty of understanding and interacting with data. The other is agent knowledge modeling. Here, knowledge modeling can be characterized as a set of techniques that focus on the specification of static and dynamic knowledge resources.

Modeling Curriculum Design Patterns

The emergence of design patterns for dealing with chaotic systems has applications in many fields. The real issue that needs to be addressed is when utilizing the framework to do so is appropriate. The course-development team and, in particular, the course designer should be able to utilize design patterns to assist in the modeling of the courses.

Design patterns can be used as a powerful tool in the creation of a curriculum’s plans. When used together with the application of Learning Objects Oriented Course Design, design patterns are another tool that can be applied in order to achieve more robust, flexible, and adaptive curriculum plans, to organize LOs into cohesive yet independent course structures.

The ability of design patterns to do more than document the curriculum plan and course design decisions that were made in its creation offers a degree of protection for the adaptability built into the system. Design patterns also offer another way in which the curriculum plans can be conceived and created. The addition of this layer of abstraction to the Learning Object-Oriented paradigm clearly allows for a more robust and adaptable curriculum.

---