Detroit junk yard © Ian Spanier Photography
LA Photographer Sam Frost,(http://samfroststudio.com) recently asked me how I label my images, organize my storage, and what drives I use for my storage.
Although my desk generally looks like someone locked a rabid raccoon in my office and let them run wild, I do pride myself on an organized library of images and a good back up system.
When I was starting to make the slow transition from film to digital I began by having my film scanned at my labs. At the same time, I was shooting a bit of digital here and there, but ultimately waiting for Canon to come out with a full size sensor before buying a digital camera. My negs and chromes were stored in ph-balanced sleeves organized by job name, separated into clamshell boxes by client. Figuring out a good system for digital files had its share of failures, but ultimately I came up with a pretty good system.
Initially I used a program called iView to catalog the images. This allowed me to assign keywords, client names, subject names and pretty much any other markers that would allow me to look up images at any time, with little or a lot of information. Keeping all the files on a computer was not going to work, so I stored all the files in individual job folders that were copied onto CDs, (yes CDs). I would name the CDs is-00001… then copy the same amount of information to an external hard drive. The CDs were stored in a 3-ring binder and the hard drive I kept hooked up to the computer. I had a separate portable drive to bring with me to jobs and on the road. CDs were pushed aside for DVDs, and as external drives got less expensive, I would move on to larger drives. The downside here was there was a ton of time being spent on making the DVDs, and the drives were not well protected. I primarily worked off the drives, and then went to the DVDs or CDs if there were any problems.
Then Drobo came out. Drobo is a self-replicating hard drive system, essentially a RAID system, which can easily handle up to 16TB of data. This means that if one of the 4 (or now up to 8) drives crashed, the other drives would protect the data and you’d only loose the drive itself. I now have two Drobo units, one with data from 2008-2011, and the other with 2012-13. The Drobos use Western Digital Enterprise class internal drives. I purchased for about $25 a SATA dock that accepts the same drives, and using them, about every 3-6 months I backup my latest work, store those drives in static bags, placed inside a Tupperware case and stored in my storage unit in NYC. Seems a bit extreme, but what this essentially does is creates a system where my files are always backed up, and should there be any problems in my office, I have files on drives in a completely separate location. All the current jobs I am working on, (jobs that clients have yet to make final image orders), also live on 2 portable hard drives that travel with me to jobs. That way, the files are always accessible, and to be even more anal, the current jobs are in three separate locations at most times.
When I dropped the DVDs I realized that naming the job folders should change as well. The sheer volume made it just too difficult to use the same system. Instead, jobs became separated by the best unique naming convention I could figure out, the shoot date. I use this system for everything now, which you will read more about below.
This is my regular workflow that I use on every shoot, whether I am shooting to CF card, tethered to Adobe Lightroom or Capture One. I have to say first off that I owe a lot to Scott Kelby’s Lightroom book; he is worth following for all his tips and knowledge, @ScottKelby
I start by naming each shoot with the shoot date, like 041213 then a “_” and some abbreviation of the client/subject. This is arbitrary and is just an easy method for me. So if my client is Danskin the job folder would read, 041213_dsk. Images would be named the exact same unless the client needs the files to be named specifically something else- like in the case of a clothing catalog where the style number is preferred. Either way, the files are numbered sequentially, with either four or five digits.
On the shoot, files are shot into the Pictures folder on the laptop’s or tower’s hard drive. I label this folder RED to signify it’s the only folder and it needs to be backed up. During and after the shoot, the files are backed up to the portable drives. Once all the files are copied, I change the RED folder to ORANGE to signify that it has been backed up to an external drive. On the external drive I have a folder for new files which is labeled RED, one labeled BLUE for folders that I have moved to the Drobo, GREEN for lores files for clients, a second RED for files to retouch (as they are not backed up in that state), and finally a GREY folder for folders that are backed up to the Drobo and cataloged.
When I return to the office after the shoot, I copy the folder from the external drive to the Drobo, and change the folder to BLUE moving it into the BLUE folder then import it into my Lightroom catalog on my office desktop. Once I have finished, I change the folder on the Drobo to GREY.
I create a new Lightroom Catalog by calendar year to keep the catalog working efficiently.
Screen shot of my portable drive color coding system.
I keep these folders on the portable hard drive until the job is completely closed. I mirror that portable drive with my other portable drive- just to be safe when I am on the road.
I cannot stress the importance of metadata enough. When I am creating or my digital tech is created a new catalog or session, my information is immediately being embedded into the files. The job name and key wording are all being done on set. This includes the client’s name, subject’s name, gender, ethnicity, location and so on. Many of my clients take the RAW files right from set, or I send a drive with images to them soon after the shoot, so this is not only important for the client then has files with this same information, but also for myself the future.
Do you really think you will remember every subject’s name? What if you get a call and someone wants to buy an image from two years ago and you have to go through thousands of images to find the subject of the photo? Time will be drastically saved if you can filter more specifically to the images you need. Recently I was asked to pull three years of images from six separate countries that needed to be separated by specific locations. Thankfully because I had the embedded metadata I was able to do this in no time at all.