Although providing open learner models to teachers and learners has proven effective, building accurate learner models remains a very complex task, partly due to the large amount of data that must be analyzed. We propose a method for specifying an open learner model at the conceptual level. This model re-uses constraints or indicators already existing in the environment, in order to expedite student modeling. Constraints and indicators convey information about student knowledge that can be easily extracted. This method is independent from the domain and from the learning environment; it exploits the interaction tracks after a learning session. It has been implemented and a pilot study has been conducted with an indicator-based environment for Java programming. Initial results show that this modeling method can provide relevant information about learners' knowledge in an efficient manner.