dedupe VS ffupdate

I wrote https://github.com/Gumnos/dedupe which sounds like it might be useful to you. It's faster than several of the alternatives I've found (many run the checksum across the whole of every file, this uses the file-size as a first-line discriminator, and only if the files are the same size does it go to the trouble of checking the checksum of the files). I designed it for creating hard-links in my media collection, but in the --dry-run mode, it should emit the file-names allowing you to pass it to xargs to remove them if it looks copacetic.

You can use my dedupe.py script with the dry-run flag (-n) to find all the duplicates on your drive. If you run it without the dry-run flag, it will attempt to make hard-links so that each file exists only once on the drive with multiple hard-links to the underlying file. It should be pretty fast, only needing to checksum file-content in the event that files have the same size (several other such deduplication methods work by checksumming every file on the drive which can be slow).

a file-deduplication utility that hard-links duplicate files to save space (our family photo gallery gets pics put in multiple albums for various audiences, so I can cut down on a lot of duplication with this)

You by no means have to use the snap. I use a shellscript: https://github.com/charmparticle/ffupdate

I prefer to just use ffupdate and tbupdate. Firefox releases binary compatible browsers for linux, why not take advantage of it?

ffupdate is a script which will update firefox, either vanilla, beta, nightly, or devedition. I made this because debian does not provide the latest firefox, and downloading and installing firefox by hand, while better than compiling from source, is tedious. This script automates the whole process of installing and updating firefox, and it can also install and update thunderbird. I have since learned that lots of other people wrote scripts to accomplish the same thing, so my scripts get lost in the shuffle in github.