Organizing
Digital Images

In the old days when we produced slides, photographers would often store their transparencies in plastic sleeves in file folders with labels that matched the subject matter to be stored, they created labels for each of their slides and then stored about 10,000 slides per filing cabinet. The cabinets were typically organized by location, client or subject to allow the photographer to rapidly find what he wanted. Photographers used light boxes to view each slide when returned from processing and would choose the ones to keep and discard the rest. With the advent of the personal computer, people created databases to store the information about each of the slides including where it was located in the files. While the system did work, there were lots of issues. I don’t know of any serious professional photographer with a large image collection that had entered their whole collection of slides in a database. The process was just too time consuming to ever get completed. Most could find images by knowing how their files were organized when they needed them and they knew roughly where the slide they were looking for was located and then they would leaf through the folders until they found what they needed. The process of picking the keeper slides was laborious and time consuming, not to mention down right boring. Labeling the slides was also a major challenge requiring a lot of time. The other serious issue was that there was essentially no backup for the images. Slide dupes were notoriously low quality and while you often can shoot multiple frames of a subject, the best shots were typically once in a lifetime action shots that you were lucky just to capture one. The hours spent in selecting, cataloguing, labeling, storing, and retrieving images were incredible and the risk of losing all your images was very real.

Fast-forward 10 years and a lot has changed, but a lot is still the same. For many, there is no longer film to store. Digital cameras name files following a standard process and the early process was to simply move the files into the computer. In essence hard disk drives, or CD’s replaced the filing cabinets of the old days.

We still have to deal with the process of selecting, cataloging, labeling (naming), storing, and retrieving our images. Fortunately, now we have a means of backing up our images, but this is another process that needs to be dealt with.

The file naming standard used for digital cameras, is limited to 10,000 file names, which sounds like a lot, but for a professional photographer or a serious amateur isn’t that many. As the camera starts renumbering at file number 9999 simply copying the files into a common directory once this number has been reached risks replacing files you wish to keep with new ones, not a good idea. So file naming needs to be dealt with in a systematic way.

In order to pick keeper images, after a shoot a photographer needs a systematic method of reviewing each one to determine image sharpness, composition, etc, in order to pick those images to keep and those to discard.

Once the keepers have been selected and the others discarded, the images need to be organized in some way to allow the photographer to select them later when they’re needed.

Fortunately a whole set of software tools has emerged to enable a photographer to deal with these issues.

In order to get around the 10,000 file name limitation, I rename all the files when I transfer them into the computer. On my Mac, I use an application called Photo Mechanic (it has both a Mac and windows version, but there is also other software on both platforms that will do the job as well) to do the transfers. I use a file renaming option to name the files for the date and time of the image capture using data embedded in the file. In order for computer sorting to work correctly, I name the files in the format year-month-day_hour-minute-second-subsecond.xxx. I start with the year so if I sort the files, I don’t intermix years (04-30-06 would come before 05-01-05 although it was taken nearly a year later). The software also creates directories for each download date and places the files in those directories. Originally I had created a directory / file name structure that included the subjects and an index number. For example I had a birds directory, under that a wading birds directory and then the files included an acronym for each subject, like GBH for Great Blue Heron and an index number to keep each unique, for example GBH-00001.NEF was stored in the Herons directory, which was in the wading birds directory, which was in the birds directory. The problem with this approach is that it is very labor intensive to go through each file, appropriately name it, and then place it in the correct directory. This has to be done almost immediately for each image taken in a shoot or the whole directory structure breaks down. I found that keeping up to date with this work was almost impossible and I ended up with directories of files that still needed to be placed within their proper directories for long periods of time like ever growing black holes of images that were difficult to find. This way I have a well organized set of images that are filed by when they were taken, each shoot can be quickly and easily located.

Once I’ve downloaded the images from the memory card into the computer, Photo Mechanic also allows me to quickly blow the images up and tag the ones I want to keep. The images are displayed in the form of thumbnails in a browser and each image can be double clicked to look at a larger version of the image to judge image sharpness. Again other applications can perform the same function.

But how do you find images if you don’t name them using the subject you ask? I embed information called metadata into each file containing such information as key words, copyright information, my name, etc. There are standards for this data and it can be written and read by many applications. The vast majority of this information is common for every image in a download and fortunately Photo Mechanic (and other such software) allows me to create a template to be applied to each image as they are downloaded. For unique situations I need to refine the keywords later, but it’s less onerous if the common elements are dealt with first and if time doesn’t allow, I still have usable information now embedded into the image. I then “import” the downloaded directories into a Digital Asset Management application (DAM) which allows me to search all the images in my collection, regardless of the storage device they are contained on, using the metadata embedded in the images. In my case I use iview multimedia pro (ivmp), but again other applications are available to do the same thing. Ivmp also allows me to refine the key wording of images, sort images by multiple criteria, rate images, create contact sheets or customer specific web pages. The good news is that the metadata follows the digital files and can be read at any step in the workflow, even when being edited in Photoshop. In addition, the spotlight feature within the Mac’s OS X operating system is also able to search for files with some of the metadata that is embedded as well.

Finally the issue of backup, it’s not if you’re going to have a problem, it is when. I carry the backup process to the extreme. My Mac has what is called a mirrored RAID disk array, which copies files saved on it to two independent disk drives, if one fails, everything still exists on the other. This is my first line of defense in case a hard disk fails. I also keep an external backup. This not only deals with the possibility that multiple drives will fail at the same time, but also deals with the possibility I delete or otherwise destroy an image on the array, as both copies will be destroyed at the same time. Finally I burn a DVD with all the same images on it. I also send a copy of my best images on DVD out of state as a final guard against a geographically based disaster.

July 2006