[Magic Man]

I was not pushing, I was merely replying to an answer in the same tone that it answered my previous post.

I understand perfectly well that "Filesystems manage files. Calibre manages books." but then Calibre manages its books using a filesystem, it doesn't store them in a db or other container so the point is moot.

I can understand perfectly that if there was a requirement two have more than one book directory with exactly the same name under the same author directory then there would need to be some form of unique identifyer in the book directory name to identify it, hence the id value, my point was that wouldn't you want to identify the difference yourself so that such a difference was human intelligible in the library? I.e. name them differently to identify them to your eyes in the library as well as enabling to program to tell them apart?.

My point was that since the directory structure has to be itterated over to retrieve these identifiers anyway, why not simply identify the records by name alone with a corresponding lookup value in a root level .ini or .db file?

No worries.