Researchers have long been fascinated with the phenomenon of lurking and free riding in knowledge sharing. This interest has led to the investigation of which factors drive decisions to contribute to a knowledge exchange as opposed to only exploiting the information in such exchange. Many studies have specifically focused on identifying the extrinsic and intrinsic motivational drivers for knowledge sharing in communities of practice by administering user surveys on behavioral intention, expectations, and satisfaction with the community. Our analysis is different from prior studies in that it does not look at expectations of reciprocity and other individual characteristics. Rather, it extracts and analyzes interaction data and, then, it groups such data based on factors like geographical location and related cultural background. This study adopts known models of national culture and relates them to social interactions using a large dataset mined from an online community of practice. The results show interesting deviations from the literature, which may be limited to the specific community of practice (programmers sharing coding knowledge) or may guide the design of open innovation systems that support knowledge sharing. This paper presents the first step on why and how to conduct such studies and suggests open questions for future study.