Demands of Modern PLEs and the ROLE Approach Uwe Kirschenmann, Maren Scheﬀel, Martin Friedrich, Katja Niemann, and Martin Wolpers Fraunhofer Institute for Applied Information Technology, Schloss Birlinghoven, 53754 Sankt Augustin, Germany {uwe.kirschenmann,maren.scheffel,martin.friedrich, katja.niemann,martin.wolpers}@fit.fraunhofer.de http://www.fit.fraunhofer.de Abstract. We present basic concepts and an outlook on current ap- proaches and techniques of personal learning environments to point out their demands, focussing on recommendations in self-regulated learning scenarios as a major basic functionality of PLEs. In the context of the ROLE project, we explain how we plan to meet these demands by using user observations stored in the format of the contextualized attention metadata schema. Keywords: personal learning environments, recommendations, attention metadata. 1 Introduction Personal learning environments (PLEs) become more and more popular in the ﬁeld of technology enhanced learning (TEL). In this paper, we refer to a PLE as a mix of applications (possibly widgets) that are arranged according to the learner’s demands. A short example will illustrate this simple deﬁnition. Sara, a learner, wants to deepen her knowledge in calculus. As she does have some experience handling/constructing her PLE, she populates it with a widget that facilitates the use of a mathematics search engine. The widget displays important information about the characteristics of mathematical functions in which Sara is interested. The widget does not satisfy Sara’s needs; she therefore also incor- porates a specialized graphical widget that allows her to display the functions and to understand them more easily, e.g by manipulating the function graph manually. She also wants to know how to implement mathmatical functions in a programming language in order to calculate a range of values as well as some derivations; she includes a special programming course and a chat widget allow- ing her to discuss ideas with people whom she might not even know yet. The output of her calculations shall then be displayed by the graphical tool. This example can easily be extended. It is apparent that the learner in our scenario already knows what she wants but that she has no clue where to ﬁnd a M. Wolpers et al. (Eds.): EC-TEL 2010, LNCS 6383, pp. 167–182, 2010. c  Springer-Verlag Berlin Heidelberg 2010