When we refer to file-based databases we mean databases that we interact with/update directly (via an SQL abstraction offered by the JDBC driver) but in essence we just read/write to a file directly. Example would be SQLite
No, CSV is a comma separated values file which allows data to be saved in a table structured format.
A “flat file” database allows the user to specify data attributes (columns, datatypes, etc) for one table at a time, storing those attributes independantly of an application. dBase III and Paradox were good examples of this kind of database in the CP/M and MS-DOS environments, and the original FileMaker was a good Mac O/S example.
A relational database takes this “flat file” approach several logical steps further, allowing the user to specify information about multiple tables and the relationships between those tables, and often allowing much more declarative control over what rules the data in those tables must obey.
Database systems are mechanisms that mediate between file systems and an API based on RDBMS principles. Without getting into detail about the specific sets of rules that together make up an RDBMS, we can say that these mostly provide a way to organize and access information stored in files. The most noticeable features of databases compared with raw file-systems are structure and indexing.