Backups were actually better. Shit like DATs and mini-DATs were common. Now outside of an industrial setting you barely have any choices other than using other hard drives, trash like optical media, etc. It used to be a snap to set up a tape backup with total and incremental backups. You can still get those but not at any reasonable price.
Now one of your best bets is operating out of a data center and having RAID, but as we just saw recently with the Farms, literally the entire RAID can simultaneously fail and you're fucked.
In my workplace we've had at one point maybe 4-5 machines that had a simultaneous failure of all SSDs and all data was lost. In over a decade of experience I've only seen it happen once, it's incredibly rare, but it can happen. You should always assume your existing disks will fail at one point and have another location to backup anything important.
A rather unpopular opinion I have is that there really isn't any reason to run a form of redundancy(e.g RAID) for most home servers taking into consideration that the vast majority of home servers are mainly used to store media in a generally write-once then just read fashion, so basically just torrenting media or saving their own.
For those specific circumstances you rarely have an actual need for redundancy, and it can bring quite some negatives:
* if failure does happen you just rely on the redundancy and don't end up testing your backups often or restoring from backup
* it's extra money that could've been put into backup drives which also raises the entry level cost for someone looking to get into this
* the extra money invested in raid has people often skip proper backups, especially if they're goaded into anything more than RAID1
* the consistent writes to everything raise your chance of disk death across board. If the redundancy disk was used for backups only, its chance of dying is greatly reduced
If you have important work data or things you genuinely need redundancy for, you can just have RAID for those, and use something like mergerfs for your media.
I'm betting the linux wizards here obviously have both and test their backups and restoration processes regularly and are super happy investing in 2x(or 3x/4x) the drives they need + another set for backups, but realistically people who go for raid will just skip proper backups.
I'm not saying redundancy can't be useful, but I have seen very few people have a genuine reason to use it, and I don't think it should just be blanket recommended for absolutely everything. Having a solid backup strategy is far better to prioritize than setting up redundancy. Hard Disks die rarely, but humans are stupid every day. I bet all of us accidentally rm -rf'ed something important at one point or another.
