Command shells are pretty common and come bundled with just about every database. These are usually simple filters which allow you to log on to the database, execute commands and receive output. They can either be very simple command-line shells or nice, fanciful graphical user interfaces. Whichever the case, command shells provide access to the database from the machine on which the database is running. We will see more examples of this methodology when we discuss CGI-based web databases in part three

Socket-based middleware does the exact same thing, but over a network. Thus, with a socket-based system, I could send commands to my database in Los Angeles while I am working in Singapore by perhaps using TCP/IP over the internet to connect to my database. We will see more examples of this when we look at JDBC-based web databases in part four.

You will almost certainly want to get access to a database and its command shell just to try things out during this tutorial. Hopefully your ISP will provide a database to test, but if not, you can probably install a database like Microsoft Access on your local system just to play with SQL commands.

Of course, whether you choose a command shell or a socket based connection, you will need to know the language of the database in order to send commands. For this tutorial, we are going to focus on SQL which is actually the defacto language of databases today.

SQL comes in many flavors depending on the proprietary database system implementing it. However, regardless of the proprietary extensions, the core of the language is the same in every environment. Fortunately, for most web uses, generic SQL will be more than sufficient.

SQL protects us from the implementation of data storage and retrieval. Essentially, so long as we use the standard commands such as

SELECT * FROM database1

It is up to the database itself to figure out what to do. We don't need to know any of the internal workings of the database or how it stores data on the file system.