The other weird thing about the standard RAID conversation is the assumption that every bit on your disk is equally valuable, and that losing a single bit is as

The other weird thing about the standard RAID conversation is the assumption that every bit on your disk is equally valuable, and that losing a single bit is as catastrophic as losing an entire drive. Yes, RAID rebuilds are hard on a drive, and yes, you can get read failures during a rebuild that can cause data loss during a RAID5 rebuild. However, of the multiple TB stored on my drives, if I lose a few sectors to a read error on the rebuild, what are the chances that I will even notice? That error could be in unused space, or a movie I downloaded from iTunes, or any number of blocks that store necessary but non-personal data. I wish that there was more data distinguishing total drive failures from single block errors and from everything in-between.