J.M. Cueva Lovelle et al. (Eds.): ICWE 2003, LNCS 2722, pp. 98–107, 2003. © Springer-Verlag Berlin Heidelberg 2003 Planning: An Intermediate Solution to the Problems in Design J. Bravo, M. Ortega, M.A. Redondo, and C. Bravo Castilla-La Mancha University, Paseo de la Universidad, 4 10071 – Ciudad Real - Spain {Jose.Bravo,Manuel.Ortega,Miguel.Redondo, Crescencio.Bravo}@uclm.es Abstract. A great variety of simulation environments aim at supporting education through a modelling process. However, there are certain problems associated to modelling environments. We intend to build the solution model to every problem through plans based on intermediate languages outlined by the students. These learner plans are abstract solutions to the design problems. Our proposal consists in a tool, Plan Editor, to help the students design a domotic environment, following the intermediate language approach as a first step in the resolution of the problem. In the end, the student will be able to more efficiently simulate the plan proposed. 1 Introduction A great variety of simulation environments aim at supporting education. There are obviously multiple advantages in them. But there is not an evident conclusion on their effectiveness since, while some authors show its excellences [20] [10], others do not believe in it [7][17]. However, everybody agrees that these environments motivate the learner [1], since the achievement of the goals produces their satisfaction [5] and the knowledge adquired through the simulation is more intuitive and efficient [2]. Simulation environments can be used to promote discovery learning. The concepts of exploration and discovery greatly motivate the learner who has the initiative. However, they bear the responsibility in the learning process [8]. Here, the teacher is just an adviser in those situations in which the learners doubt about their own performance [16]. There are certain problems associated to discovery environments. To mention some, we can start with the incorrect execution of the exploratory processes of learning. De Jong finds out that only 50% of the processes by discovery are fulfilled in an efficient way. Another problem is the incorrect relation between hypothesis and experiments [13]. Wason [22] demonstrated by means of a series of experiments that students do not often follow logical thought. Finally another problem we found in this type of environments is that the students’ knowledge of the model is not precise enough. Besides, the degree of freedom in knowledge acquisition in this kind of learning does not suggest any ordered mechanism or guidelines that the students must follow. So, responsibility together with freedom of action can, in some cases, frustrate the learner, preventing the training goals from being successfully achieved.