I keep most of my personal/project files in a folder structure that 'lives' in the cloud, but has a full offline mirror on my main devices (desktop and laptop). From my point of view this is an essential quality of life thing, and I have no idea how other people survive without doing this.

The problem I have now is that I'm on linux, automating my mirroring with rclone. Rclone is great, but if a file gets corrupted during the sync process, it can't tell, which is A PROBLEM because over time with repeated syncs, corrupted files can potentially end up overwriting good files.

Normally this problem can be solved by running checksums to verify integrity. A script can make sure that everything matches up correctly, and can automatically repair things instead of propagating destruction upon all my files. Unfortunately, it seems that Proton Drive hasn't exposed a system for this yet, so there is no reasonable way for me to do these integrity checks that are a key part of any healthy synchronization framework.

The number one thing a file system should not do is destroy its files. Checksums are kind of 'the way' we make sure we don't explode things. This is really quite an essential feature.

Edit: IMPORTANT: It seems I should mention that checksums can also MASSIVELY reduce the overhead traffic on your servers. They allow a sync engine to see if a file is the same, and its metadata just changed for some reason, so then the sync engine doesn't have to re-download the thing it already has. The users want this to speed up syncs on their end, and you guys should want this to get better performance for less server load on your end.