This article was supposed to be about my new collection of Korg Nano devices, how cool they were and how it was great to have MIDI controllers that cost about the same as a toothbrush. In fact, I wrote that article. But on Christmas Eve, just before I started assembling the 14-foot tall Bob The Builder Backhoe for my 4 year-old, my laptop’s hard drive crashed.
It didn’t actually crash. Not in a cool, Fatman-link sort of way. Rather, it just stopped working in a very sad and silent way.
This forced me to consider my backup strategy. I don’t really have one. Once in a while, I’ll boot the desktop into Firewire mode and dump some portion of my laptop into one of the big drives, but this only happens “when I have spare time”.
Spare time. What a joke. About as easy to find as the spare tire on a ’73 Ford F-100 pickup truck.
So I ended up losing a butt-load of my recent writing. Luckily for me, my code is all backed up to an SVN repository (wisely located at least one continental divide away from me), and my recording sessions are on a redundant pile of hard drives, but the stuff that existed solely on my laptop has been vaporized.
Gone forever.
Ever since I was an Young IT Professional In The Great Tradition Of IBM Mainframes, the idea of a backup plan has been pounded into my head. I’m older now, which means that I’m much more capable of rationalizing stupid behavior. Not having a laptop backup since September 24th could clearly be called stupid behavior.
Reading this, I’m sure you are clucking your tongue at me - while, at the same time, you are probably hooking your Firewire drive to the laptop for a quickie backup. While that is going on, why don’t you also take a look at my latest weapon against dummy-hood: http://www.getdropbox.com. It’s a cool way to keep your work in The Great Cloud - and gives you some semblance of on-the-fly backup.
My main writing folder is now a Dropbox folder, and all my work automagically gets pushed to the Great Stash In The Sky. If only I’d done this before December 24th, you might be reading about my Personal Nano Love instead of this bit of self-immolation.
It's easy to get complacent about doing backups. Let me urge you to fight that complacency. This article might not seem like it's about audio, software or even digital media - but all of this become pointless once the bits leave the building. So let me be your cautionary tale, and try to find a comfortable way to fit hard drive backup into your personal workflow.
Now do as you've been told and get to that backup!
Categories
Audio
Read More Entries by Darwin Grosse.
Become an O'Reilly Member
Visit the Digital Media Forums
Subscribe to the Digital Media Blog Feed
Subscribe to the Digital Media Newsletter
Thanks for the Dropbox tip, Darwin. The concept of uploading just the parts of a file that have changed and storing previous versions is slick, but automatically creating another backup on a second computer the user owns makes it even cooler.
I've been using Foxmarks to keep my Firefox bookmarks synced between multiple computers, and I love it. To do the same thing with other documents will be quite a time-saver. As far as I could tell, Dropbox is free, too.
I imagine we'll eventually store a lot of our creative work in the cloud and then pick at it with a variety of smart and dumb devices. I already try to stick with programs like Adobe Fireworks and Ableton Live that run on both my (desktop) Mac and (laptop) PC so I can work on my documents in the most convenient way. But I hope we'll see more file formats that are widely compatible so that the same document can be edited nondestructively in a variety of programs.
Looking forward to hearing about those Nanos! I'm holding out for the black models.
Great overview, Stef. Automated backups are an essential part of a balanced backup diet, but laptops provide a challenge to that function. In order to back up to an external drive, you need to have access to the drive; however, if you are in process of flying, driving and training about the country (as I have), it is easy to either forget your backup drive or simply be unable to take it along.
As far as backups go, it's *essential* that they're automated – ones that you have to initialise yourself are bound to end up with the above situation where the backup's too old. Here's what I'd suggest:
For PCs, Acronis True Image Home is about the best I've found, but it's not free. There are some external USB drives that come with free backup software (e.g. Maxtor OneTouch) which work OK, but I've not found anything that compares. If anyone can suggest something else that'd be great.
For Macs, upgrade to Leopard if you haven't already and use the superb feature that is Time Machine. If you've got a laptop and always use it wirelessly, consider buying a Time Capsule, they're very good (though do the initial backup using a network cable or it'll take forever). If you're stuck on an older OS version, there are still a few options like Carbon Copy Cloner, SuperDuper! or Lacie's Silverkeeper.
For Linux, I'm not too sure about desktop apps under Gnome or KDE, but rdiff-backup works well for incremental backups from the commandline, and rsync is also worth a look.
As for data recovery of the dead or dying disks. You'll need a desktop PC and the ability to plug it directly into PATA or SATA, which may require an adaptor if it's a laptop drive. You also want another drive that's at least as big as the dying one. To try to recover the lot by cloning, use the Linux GNU ddrescue utility - it should be on most good LiveCD distributions (e.g. http://www.sysresccd.org/ ). It's not got a GUI so you'll need to be comfortable with the commandline, but it'll allow you to clone the dead drive to the spare drive and retry the bad sectors till it gets a useful result. It may well take a while, though, depending on how dead the drive is (one drive I recovered too several weeks to complete). Avoid Spinrite on dying disks, it'll just kill them quicker (it's a good tool, but not for that).
Other tools to look at include TestDisk:
http://www.cgsecurity.org/wiki/TestDisk
and PhotoRec, which will scan the raw data instead of the file allocation table to try to find file headers directly, meaning it can recover data after an accidental format or curruption of the filesystem's index:
http://www.cgsecurity.org/wiki/PhotoRec
Both are worth a try, but I'd go with ddrescue myself first if possible, then use these tools on the cloned drive, as you'll then not have to deal with failing hardware.