Post #302647

Here's the thing though, given the lack of alternatives when lacking money (back then), I'm pretty sure I would have just sucked it up and deal with the fact it takes forever to write anything. Multi-device could've just been handled by LVM+journaling as usual. The point of the filesystem would've been removing reliance on hardware errors bubbling up properly (why older blockwise RAID without integrity features (this is the part that made the good hardware cost a fortune, besides plain simple greed) is useless).