I'm going to use the latest Data Dump (September) and do some SQL queries along with it. The code for the Data Explorer (located here) explains how to use the Data Dumps with it. However, it doesn't mention whether you have to convert the XML's from the data dump into an SQL database. I know how you can convert those, I'm just wondering if it is needed in order to use the data explorer.
Yes, you need to import the data dump files into a SQL Server* database before you can use Data Explorer against them. There are some applications on Stack Apps that will do this for you, although I don't think any of them are currently maintained (and therefore might not be able to import all of the data in the current data dumps). I personally used (a modified version of) SoSlow when I last did an import on my machine.
Also, you should use the GitHub repository for the current version of Data Explorer, as the Google Code page might not always be up to date. Hopefully the documentation can be updated everywhere sometime in the near future to avoid this causing confusion in the future.
*Support for other databases might make it into the core repository at some point, but if not I'll put up a branch for it sometime when I have time, as this is something I already have locally