We define semantic density as an attribute of a part of an ethnographic corpus. It is the property of encoding meaning, in a sense identifiable by, and relevant to, an ethnographer. In the case of Semantic Social Network Analysis, the ethnographic corpus is a conversation held on an online forum, such as https://edgeryders.eu; its parts are individual posts. We explore the pattern of semantic density in an annotated corpus. Our final goal is to find algorithmic ways to determine which posts ethnographers will consider worthy of attention.