http://en.wikipedia.org/wiki/Univers...que_Identifier
Yes. As I stated in my post.
Hence the 'noauto' option.
And when you do plug the device, then it can be mounted with a consistent mount point, or from a script, which is what I do. This particular unit is normally plugged to a server for backups scheduled as a cron job. The backup script mounts the device, performs the backup and unmounts the device just prior to the script exiting.
The device in this example is formatted with as NTFS.

I still don't really get it.

So how do I use it? Could you explain to me how that fstab line helps?

Why not make a udev rule that makes a symlink called "backup" for the first partition of a drive with the correct UUID?

I want a script to automatically mount it, and then unmount it when the backup is done.

Why would you format your Linux flash drive with NTFS? Why not FAT, which has better Linux support and still works fine with Windows?

The UUID is a unique identifier for your device.
The fstab line saves you from having to make a custom udev rule, but still allows you to have a consistent mount point.
The script you will have to write for yourself to suit your own circumstances. My script simply mounts the drive, runs a couple of rsync commands, then unmounts the drive. You can find it here. http://www.linuxquestions.org/questi...1/#post3810632
The device in my example is an external USB drive. It came formatted with NTFS and contained existing backups made using Windows tools prior to being put into service with this server.