The structure, dynamics, and importance of the social network of collaboration among scientists has been already studied, sometimes yielding counter-intuitive conclusions. In this paper we investigate the role played by people who served as PC (Program Committee) members in the network formed by members of the Brazilian computer science community and their co-authors. Some characteristics of such network are compared with those reported in similar studies involving other scientific collaboration networks. As a result, we show that apart from the evidence of Milgram’s phenomenon (six degrees of separation), there is no other community with completely similar patterns (among those used for comparison). This is probably due to the unique characteristics of the target network. For instance, their members do not necessarily interact with each other in terms of co-authorship since they belong to different sub-areas of computer science. There are strong evidences that the clusters in this network are connected by non-Brazilian members. Moreover, nodes with high degrees have little connection to Brazilian authors.