I think the point is that ockam site requires a flexible layer to manage its data, which is focused mostly on .md documentation files. So Storing, updating and managing these files through a database could be quite hard and annoying, since all of them are stored across many different github repos. Moreover, all these files, have to be shown in a user-friendly way. This data from .md files, have to be converted to HTML at some point as well.

So what we have done is we keep all data on their place ( depended github repos ), grab them before build time, and generate static content based on it. Moreover, we have some mechanism to target particular branches/commits on depended repos, and control when we want to trigger our pipelines.

So, replaying directly to your question - no, building a static site not require a database. JAMStack architecture is helping us to provide and manage our specific data sources.