The Java implementation can be run by doing the following.
We have the database configured to use gdbm files (this can be changed in the
cube.database package to use your favorite DB, which shouldn't
be too hard).  Unfortunately, gdbm files do not exist on some platforms,
so if you are running on a non-Unix box you may be out of luck.
Anyways, do the following to get the cube going.

0) Decide which dimensions you would like to support.  
1) Add those dimensions to the specifications directory.  Each dimension should
   be in a separate file.  Three dimensions currently exist in specifications:
   Time, Machine, and Page.
2) Add the filters to specifications/Filters.  The filters are the regions 
   in the Incomplete Cube that you would like to maintain.
3) You must configure configure/Constants.java to your site.  The edits you
   will need to make are commented in that file.
4) Now you are ready to create a new cube.  Execute the following in sequence.

     make cleanDbs   
     make code
     make dimensions
     make count
     cd gui; java CubeGUI

   `make cleanDbs' will clean out the old cube database (gdbm files), you
   download these as part of the package, so you can just skip to the
   last step and try out the cube, it may work.  But just in case, there
   is no harm in recreating the cube.  `make code' recompiles the classes
   as needed (all the classes should be there).  `make dimensions' will 
   read the specification files and create the dimensions.  `make count'
   will then populate the cube.  It reads the accesslog in logFiles/access_log.
   Finally `cd gui; java CubeGUI' will fire up the GUI. 
