Nowadays, flexible capturing techniques allow the creation of datasets rich with information on multimodal conversational behavior. Such data consist in non-verbal and conversational features playing an important role in the interaction strategies and are used for the development of interaction models. In this paper we will focus on the representation of the interactional behavior from the multimodal conversation analysis perspective while we'll discuss the need of a data ecosystem to support the development of applications in the field of multimodal interaction.