I recently read that to minimize the risk of corrupted Scrivener database (library) backup files, it’s advisable to enable ZIP compression for Scrivener backups.
If you use Scrivener, do you know if this is accurate? I’ll cross-post in the Scrivener forum.
Here’s what Tech Support said on the Scrivener forum:
Specifically, ZIPping a backup puts the entire Scrivener project structure into a single file. (It’s actually a folder, with subfolders and potentially hundreds of component files.) This prevents “sync” operations from transferring parts of the project independently, which can lead to a range of issues. It also prevents accidental modification, since you have to unZIP the backup in order to edit it.
It takes half the space, which is reason enough for me. In any case while my projects are in synced folders (I use Resilio), I have it save backups locally so sync problems with backups is not an issue. The local drives are backed up as well , but losing the backups of the backups has never been an issue.
My backup routine mitigates the dangers of a single point of failure. I back up to two external drives and to Backblaze. I also periodically export all of my work as Markdown files to an archive folder, which is included in my backups.
And, Scrivener’s backup settings “encourage” automatic backups, and many of them. The bigger risk is putting those zipped backup files on a sync service and for whatever reason those zip files disappear (accidental deletions), or whatever, the sync service pushes them to their servers and … poof, so-called backup gone.
That is a possibility, but as I said, I have two weekly external backups of all of my files and a constant backup with Backblaze. I believe I’m as protected against loss as the average person can be.