Duplicate files

What this means

A duplicate is not simply “a file with the same name”. It is a file with the same contents in more than one place. That distinction matters in both directions: two unrelated files can share a name, and the same file can be saved under two different names.

Why it happens

Duplicates come from a small number of habits and events:

• Importing twice. Photo libraries grow quickly when an import runs over files that were already imported.
• Downloading again. The fastest way to get a file you cannot find is to download it a second time.
• Duplicating before editing. Duplicating a document so the original is preserved, then keeping both after the edit.
• Restoring from backup. Restoring into a populated folder produces a second copy of everything in the backup.
• Cloud sync. Services that keep local copies can reintroduce files you removed locally.

Why filename matching is not enough

Two failure modes make filename-based detection unreliable:

• False positives. Invoice.pdf from two different suppliers is two different documents with the same name. A name-based tool reports them as duplicates and invites you to delete one.
• False negatives. IMG_4821.jpg and holiday-2019.jpg can be byte-for-byte identical. A name-based tool sees no duplicates at all.

The Duplicate File Finder compares contents, which avoids both.

Deciding which copy to keep

Inside a group the contents are identical, so the decision is about location. Keep the copy in the folder you actually look in, and remove the copies in staging areas.

Two things to check before removing a copy:

• Does an application expect it at that path? Application support files are often referenced by path. An identical copy elsewhere does not help the application find it.
• Does anything link to it? Aliases, symlinks, scripts and other documents can all point at a specific file. Removing that path breaks the reference silently.

If either is unclear, keep both. The space is recoverable; a broken reference is easy to miss and annoying to repair.

Keeping it from coming back

Duplicates are produced by process, not by accident. Changing the import habit and the restore habit removes most of them at the source. A periodic pass over Downloads and the photo library catches the rest before it becomes a storage problem.