Some nice and sane words about backups and photography.
I think one reason we tend to save everything is because we’re still new to this, we don’t really have a historical roadmap on what to keep and what not to keep. And because it’s so “easy” to keep everything, we do, even though in reality, the logistics of finding stuff later gets so complicated that in many cases the images might as well not exist.
Of course, this isn’t a new problem. I know people who kept every negative, every print, every scrap of paper. And yeah, it’s not much better than throwing it all out, someone has to go through it some day. It’s one reason I’m trying to become more thorough, rigorous and complete in my meta-data and editing, to save someone else the choice of plowing through things for the gems or throwing it away…
In a moment, Iâ€™m going to go over my backup strategy, but before we even get there, I honestly think I might be backing up too much. Hereâ€™s what made me start thinking like that. Terry recently did a portrait shoot where he took 710 photos during the shoot. His subject reviewed the images in Lightroom, and choose the shots she liked (around 70 initially, then she narrowed it down to her favorite 5 or 6). Then Terry picked his favorites, and he chose 5 or 6.
So, what do we all do next? Thatâ€™s right, we back up all 710 photos, even though the subject has already said, â€œI only like these 70.â€ She looked at them all, told the photographer straight up, â€œI donâ€™t want any of the other 640 imagesâ€ but we back them all up anyway. Now, Terry asked me, â€œWhat are the chances that she is going to come back some time in the future and ask for one of the ones she didnâ€™t like? Right. Slim to none. Yet, we still store â€˜em, and watch them eat up our drive space, and add more complexity to our file management. Like Terry says, â€œThose 640 images are never going to see the light of day. I donâ€™t have any use for them. She doesnâ€™t have any use for them, but Iâ€™m backing â€˜em up anyway. Why?â€
In my case, I break things down into three categories: keepers, okay, and dings.
Dings get thrown away; no reason to keep a bad photo, period. Of course, sometimes you have to slow down and realize that it may not be what you intended, but it’s not a ding. On the other hand, if you do throw it out, would anyone really care? just to keep it in perspective
My keepers (basically, stuff I rate 3 stars or better) end up in my primary and end up backed up in six different places (no, seriously: my primary hard drive, my time machine OS X backup, two different bootable backups via SuperDuper!, one on my Time Machine disk, one on a bus powered firewire drive I carry on trips, and on S3 as my catastrophic backup). The Time Machine and SuperDuper backups to that drive are automatic, the one to the bus powered is manual and I generally update it weekly.
that would be the “70” photos in the group above, FWIW. the other 640 images go into a secondary firewire drive. In my scheme, they get backed up to the Time Machine disk, so they’re in two places. At some point, that’ll change, and I’ll shift that to cloning them to a 2nd drive stored offsite.
my Time Machine disks get cloned about every 4-6 six weeks, and those clones are stored offsite.
In my current evaluation scheme (which keeps changing), that 640 images would be split into 1-star and 2-star ratings. 1-star are “perfectly okay but not interesting or boring” images; nothing wrong except they don’t really work, and 2-star, which are images that on later edits of the 3 star, “lose out” to better, similar images.
For a recent shoot at Yellowstone, for instance, I took something like 150 images of Old Faithful during an eruption. throw out 20 as technical dings (“autofocus wigged”, “damn thumb in the way again”). 70 got into the 1-star folder. 60 go into the 3 star folder. When I re-edit the 3-star, I’ll end up tossing 40 or 45 of them down to 2-stars as being too similar to other photos to be worth dealing with further, leaving me with, say, 15 photos in my “primary set”. (for those curious, 4-star and 5-star ratings don’t happen here, they exist as subsets of “best of” and “killer” photos that I deal with later in the workflow.
But then Terry brought up a good pointâ€”how often do you really need 20GB for one client (or for one wedding)? He pointed me to an 8GB USB Flash drive, for only $29. If you only need 4GB of storage, you can get one for just $13.95. Heck, you might as well get two, and have two back-ups that hardly take any space at all. You could put them in a tiny zip-lock bag and staple them to your copy of the contract for the shoot. This changes the whole situation pretty dramatically; now its not eating up your main storage; youâ€™re off-loading the finished job to USB drive (or external hard drive) dedicated to that shoot (and paid for by the client).
Mistake. those aren’t archival. You could well go back in 2 years and not be able to read some of the data. or any of it. This is just asking for problems. (you might want to burn an archival DVD for a baggie for the contract, but I wouldn’t build my library around this).
Instead, My process is that about once a year, I buy a new hard drive (500 gigs, good quality, USB 2.0 — about $150), copy my libraries out to it. That stays in my home as backup. The previous year’s version (reformatted and refreshed to the current data) goes offsite. The N-2 version gets reformatted and used for something non-archival. More convenient AND cheaper with greater storage than a dozen of those non-archival flash drives, and every year the files get copied to a fresh drive and fresh media to minimze the chances of being unreadable or alpha particles or whatever. And if something comes along and replaces USB and/or hard drives, then for the annual refresh, the copy is to that new media/technology, thus avoiding the “damn, I have nothing that reads 9track tapes” problem. Or at least minimizing it. An annual refresh might take two or three evenings for a large library, but it’s maintenance well worth scheduling and budgeting for.
Ultimately my collection will get large enough to span multiple drives, even really big ones, but this plan scales pretty well, and at some point, I’ll actually decide I don’t need those 1 star images any more and trash them; but for now, they’re there, just in case. But not clogging up my main workflow, because that isn’t time or cost effective, adn slows down my “real” work…