Yes I don't think you can/should rely on TM only. It's convenient to keep running in the background but you'll still want to do manual backups to multiple places (external disks, Dropbox, Github etc)

Manual backups fall by the wayside in every environment I’ve seen them used. You are a better person than me and those are me. I’m a big fan of backblaze for this reason.

As an aside, I’m surround by people who consider cloud syncing to be a backup, and just yesterday witnessed someone in the grief stage when they deleted a load of photos to save space because they thought they would remain in the cloud. They had even clicked through the “delete delete? Are you sure?”.

To be fair, Dropbox works pretty well as a backup (with the ability to restore deleted files via the web UI), as long as you catch it in the 30 day window.

When I worked there, I was surprised to learn that they also often act as mitigation for ransomware attacks (they could roll back your account in time if you contacted CS and explained your situation).

Usually, after I complete a major project milestone, or do a lot of housekeeping (computerkeeping), or just when I feel like it's too been too long since the last manual full disk snapshot, I tend to do one.

I do this with my firewall. I’ve been caught out by making a seemingly minor change before saving a config then breaking everything.

It’s almost been enough to get me to have a test LAN and real one so as to maintain household tranquility.

Doesn't Backblaze delete copies of volumes that you haven't accessed in a certain amount of time?

Yes. It’s 30 days. Way too little imho.

By snapshot I mean filesystem snapshots as implemented by zfs and btrfs. So functionally they're snapshots, but space usage wise they're incremental backups.

Actually it's sort of halfway between ...

Which is to say, a block-based snapshot is very efficient - your snapshots only take up the space of the blocks that have changed - regardless of which files those blocks make up.

A hard-link based snapshot scheme is also fairly efficient for most use-cases - the snapshot only uses space equal to the files that have changed - even if those files only changed a little bit.

For most use-cases, I find the difference is negligible. However, if you have large files that change by small amounts, a block based scheme (a la zfs) is much more efficient.

I’ve seen too many TM corrupt itself (wifi). It’s nice for the TM “feature”. Quickly going back. But it’s not reliable as a backup solution.

Also, at a certain point it just can’t keep up with the changes.