RDF is used as a central building block for the Semantic Web. Considering providers of learning resources, it is commonplace to store and exchange meta-information in XML rather than RDF. Instead of transforming meta-data artefacts into RDF, we integrate these meta-data by translating users' queries issued against RDF into queries over XML meta-data. We demonstrate the applicability of our query translation method in a concrete application scenario taken from the educational domain.

Creating learning objects (LO) is expensive, so reuse is very important. One possibility is dividing larger packages into small LO, allowing reuse through new assembly. However, this always requires some “glue ” between the individual LO to create a new coherent, complete, and larger one. Another possibility is creating a single larger course and deriving various smaller or specific versions from it automatically. These are then complete, or at least much easier to finish, and even matching in look and feel. Although not universally suitable, often for instance a general overview or a detailed sub-topic can be extracted. The only necessity for this is annotation of smaller parts according to an arbitrary taxonomy. According to this taxonomy the LO is then filtered to produce a subset con-taining only specifically annotated elements. This paper describes a tool for deriving such subsets through convert-ing taxonomy annotations into XML, allowing the use of XPath expressions for filtering.

This paper describes a content object and metadata repositories model, which provides an alternative to search and to access repositories developed in Colombia and Latin America. Along with the repositories model, an initial design of a decision support system for repositories is presented. In order to preserve the autonomy of each repository, a common web services interface is considered. This interface can be implemented on different software platforms or operating systems, and it also can be integrated within existing solutions. The repositories that accomplish with the services interface are based in IMS DRI 1.0 and SCORM 1.2, and they will have an internal catalogue that manages the content objects, their associated metadata and other catalogues for the development of distributed operations. By the other hand, the decision support system is a tool that provides the information required to evaluate and to control the quality of both the services and the resources offered by the repositories, among other parameters. The development of the decision support system will require technologies such as Data Warehouses and On Line Analytic Processing. At a technical level, this document describes a relational database model which allows manipulating the learning object metadata stored in a content repository, no matters if the database engine supports XML or not. According to the relational model, the tables, their functional description and their relationships are presented. Finally, the dimensional models of the decision support system and the web interface offered by the current repository implementation are described.

Abstract:- Designing a Learning Management System (LMS) is a combination of pedagogical design, usability and information architecture. E-Learning systems should have intuitive interfaces and clear information design, allowing learners to concentrate on learning and course developers on learning strategies, thus making e-learning more efficient. Hence, effective LMSs should include sophisticated and advanced functions, yet their interface should hide their complexity. In this paper we present the key features of an experimental LMS which follows recent usability techniques in order to accomplish simplicity. The usage of computer file management as a metaphor for management and navigation between Learning Objects (LOs) along with a clear and unambiguous user interface allow users to focus on their learning activities without excessive hassle.