I'm running Python 2.3.4 and using distutils ( 
__revision__ of 1.58 2003/02/19) on WinXP SP2 to 
create a windows .exe installer. Both installation and 
removal of the package result in a residual file on the 
root of C: drive. This seems to be directly related to the 
use of the windows install-script which is provided with 
the --install-script switch.
 
The residual file has a naming convention of sxxx 
where 'xxx' is any alphanumeric combination. There is 
never an extension on the file name. On installation, the 
residual file contains the same text the install-script 
prints in the install screen window. On removal, the 
residual file is empty (the install-script does not print 
anything when removing.)

Users attempting to execute the installer from a 
network drive experience installer crash as they do not 
have file system permission to generate the residual 
file..