[ddrum001]

You're correct that it's very difficult to balance between giving an overview and over-simplifying. We intended this to be a starting point, knowing that we'd have to make a trade-off between including more details and giving a concise overview.

For NoSQL specifically, it's difficult to put every database in only a few categories. We discussed that Cassandra and HBase are "maps of maps" in the details, and we definitely didn't want to imply that they have the same model or use-cases as Redshift. Perhaps our next iteration will explain the "column family" more and include a separate category for the databases that were inspired by the Big Table data model.