• The Adapters leave the original learning environments unaltered for the most part. They provide the interface for interaction between the learning environment that the adapter is attached to (using the API of the learning environment or some other available way to access its internal data) and the other components used within an integrated learning scenario that includes multiple tools.
• The Composite Data Structure provides a central access place to arbitrary parts of the data to be shared between the learning applications. The hierarchical structure not only is just a structuring principle from computer science but also allows representation of educationally meaningful information at different abstraction levels and scale. Additionally, a subscription mechanism for parts of this data structure is provided. This mechanism uses notifications to inform registered learning environments about changes in shared data (parts), thus avoiding inefficient communication via active polling processes. Technology wise, the Composite Data Structure is the key to keep development costs low: without it, adapters for all possible pairs of tools might be needed. For tools, this would amount to adapters versus adapters when using the Composite Data Structure. From an educational point of view, a hierarchical tree data structure is often adequate—e.g., to represent learner models (with different skills), workspace contents, or user actions in a system (indexed by user and system component).
• The Learning Environments use the functionality of the adapter to get access to the data elements they are interested in for internal use. The processing of the data (i.e., the educationally meaningful interpretation of the exchanged learning objects) is fully encapsulated within this component.
• A. Harrer is with the Catholic University Eichstätt-Ingolstadt, Ostenstr. 14, 85072 Eichstätt, Germany. E-mail: email@example.com.
• N. Pinkwart is with the Clausthal University of Technology, Institut für Informatik, Julius-Albert-Str. 4, 38678 Clausthal-Zellerfeld, Germany. E-mail: firstname.lastname@example.org.
• B.M. McLaren is with Carnegie Mellon University, Human Computer Interaction Institute, 2617 Newell-Simon Hall, 5000 Forbes Avenue, Pittsburgh, PA 15213-3891, and also with the Competence Center for e-Learning, Deutsches Forschungszentrum für Künstliche Intelligenz, Stuhlsatzenhausweg 3, 66123 Saarbrücken, Germany. E-mail: email@example.com.
• O. Scheuer is with the Competence Center for e-Learning, Deutsches Forschungszentrum für Künstliche Intelligenz, Stuhlsatzenhausweg 3, 66123 Saarbrücken, Germany. E-mail: Oliver.Scheuer@dfki.de.
Manuscript received 24 Jan. 2008; revised 3 Nov. 2008; accepted 4 Nov. 2008; published online 13 Nov. 2008.
For information on obtaining reprints of this article, please send e-mail to: firstname.lastname@example.org, and reference IEEECS Log Number TLT-2008-06-0051.
Digital Object Identifier no. 10.1109/TLT.2008.18.
Andreas Harrer received the PhD degree in computer science from the Technical University of Munich with a doctoral thesis in the area of CSCL (intelligent support of collaborative learning interactions). Since Winter 2007, he has been an associate professor in informatics (computer science) at Catholic University Eichstätt-Ingolstadt. He was with the Collide Group, University Duisburg-Essen, from 2002 to 2007, with a research line of analysis and intelligent support of group learning as well as software engineering principles for educational systems. He was a speaker of the special interest group "Artificial Intelligence and Education," the European Research Team "Computer-based Analysis and Visualization of Collaborative Learning Activities." He is a member of the steering group of the European Research Team "Computer-Supported Scripting of Interaction in Collaborative Learning Environments" in the "Kaleidoscope" Network of Excellence and has been involved in numerous international projects and collaborations.
Niels Pinkwart received the PhD degree from the University of Duisburg-Essen in 2005, with a thesis on "Collaborative Modeling in Graph-Based Environments" and a postdoctoral degree from Carnegie Mellon University. He is currently with the Clausthal University of Technology, where he currently leads a research group that investigates collaborative systems with their software architectures, design patterns, user interfaces, and usage by humans. A specific focus of his research is set on applications in the domain of educational technology, particularly on distributed and collaborative software systems that provide intelligent support to students in order to help them learning. In this field, he has published more than 60 referred papers in national and international conference proceedings and journals.
Bruce M. McLaren is currently with Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI), Saarbrücken, Germany, as a senior researcher, and with Carnegie Mellon University (CMU), Pittsburgh, as a systems scientist. In both institutions, he is engaged in research of human learning and educational technology. He does his research within the Competence Center for e-Learning at DFKI and within the Pittsburgh Science of Learning Center (PSLC) at CMU. At CMU, he comanages a team of eight programmers and research associates in the development and enhancement of the Cognitive Tutor Authoring Tools (CTAT). His research interests and experience include educational technology, collaborative learning, intelligent tutoring, case-based reasoning, and artificial intelligence. He has more than 50 publications in journals, conference proceedings, workshops, and symposiums.
Oliver Scheuer received the Dipl-Inform (master's) degree from the Saarland University in 2007. He is currently a PhD student in the Competence Center for e-Learning, Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI), Saarbrücken, Germany. His research interests include the application of machine learning, online data analysis, and feedback techniques to collaborative learning systems.