Machine learning is now well established as a powerful tool for solving ecological questions. Yet, it is not a trivial task to develop an interdisciplinary approach involving experts of very different scientific fields, but also students at the interface. From our experience, the likelihood of developing long-term collaboration increases with the ability of experts to develop a common language in order to understand each other. To illustrate this, we will discuss several keywords, such as “interdisciplinarity”, “methodology” and “explainability”, which can hold different meanings depending on the perspective of a computer scientist or an ecologist. This discussion will be open to the audience and serve as a conclusion of the symposium "Machine learning for ecological images"