Folders aren't enough. Find file isn't enough. I wanted the database so that I could use my own organization system including tags (fields) like draft status, author and countless others I have I mind. But even then, folder structures assume that something belongs to only one family.
Also, I wanted to be able to do everything from one interface.
However, as you said, I may have to give this up and use file links. It's probably what I will do.
I just thought of something, my main resistence to using folder structures (or anything based on the normal file system structure) was the inherent limitations to it. As in, file names have to be under a certain length, and folders imply that something belongs to one category. MySQL doesn't care if I have 10,000 or more entries which are all duplicates. It doesn't care if I have 10,000 entries which all have the same "title" field, but the OS does care if a file has the same name in the same folder. Maybe I could write the program to automatically assign file names as a unique number. If I did that, I can't imagine being so prolific that I would create too many files to hit the file name limitations. It would bypass trying to think of a way to name different versions of the same story in a descriptive way, because I could use the fields in the database as my descripters. Then I could also just write a short backup script to tar, gzip the files.
It's not what I envisioned, but I think it would work.
Thanks for the input.