1. Do subdatasheets assist in normalization or just make the data harder to
maintain?  

2. In a drawer of 50 files 20 of the files contain other files (subfiles).  
To create a normalized database, would you suggest I create a table for each
of the files that contains files?  I have already created keys to identify
each record whether it be a "parent" file or a subfile.  Creating new tables
for each file that contains files seems redundant and prohibitive for easy
data entry in the future.  

Currently, I'm simply trying to convert a directory of files in Word to a
searchable database in Access, but I expect it to grow over time. Thus,
while keeping normalization in mind, I also hope that whatever I end up with
will be easy to update, add records to and query.

Thanks to everyone for their thoughts