Oracle has to keep track of all the tables, views, constraints, indexes, triggers, and other objects you create. To do that, Oracle needs a place to store the information. This repository of information about your database is referred to as the data dictionary. Whenever you create a new object, such as a table, Oracle stores all the information about that object in the data dictionary. Modify the object, and Oracle modifies the data dictionary. It follows, then, that if you want to know anything about your database, the data dictionary is the place to go.
The data dictionary is a set of tables owned by the user SYS.
The structure of these tables ends up being fairly complex, and much
of the information isn’t stored in a user-friendly form. You probably
do not want to query these tables directly, and unless you have been
given access to log in as user SYS, you won’t be able to see them
anyway. To help you out, Oracle provides a set of data
dictionary views. These views have names that are easy to
remember. The column names used in the views are also easy to remember
and use a consistent naming convention. Data dictionary views exist
for each different type of schema object, and they present information
in an easy-to-understand form. For example, if you are looking at a
date column, the
view will tell you it is of type DATE. The underlying data dictionary
table, which happens to be
sys.col$, will tell you the ...