A number of researchers in computer-mediated communication have pointed towards its potential to stimulate learner participation and engagement in the classroom. However, in many cases only anecdotal reports were provided. In addition, it is unclear whether the pedagogical set-up or the technology involved is responsible for changes in learner attitudes. We have conducted bilingual MOO (object-oriented Multiple-User Domain) projects based on tandem learning for the past three years, and collected extensive learner data on native-speaker/non-native-speaker discourse during that time. In this paper, we will look in detail at issues such as topic negotiation and initiation, both in the learners L1 and L2. For instance, some researchers on native-speaker/non-native-speaker discourse have claimed that native speakers initiate the majority of topics, and that non-native speakers preferably initiate a topic by using questions. An analysis of learner data will be compared to existing research on face-to-face native-speaker/non-native-speaker discourse. This comparison provides us with an empirical basis to assess whether a combination of tandem learning and a synchronous communication technology like the MOO can support learner participation and engagement in the language classroom. We will also discuss whether the technology or the pedagogy is responsible for any differences in learner engagement.