Controlled Vocabulary

Filenames as a Strategy to Managing Your Image Assets

The most important thing that a filename can do for your image collection is to provide a form of unique identification ( or [UID] ) for each digital "asset." Having a system that ensure that each filename is unique is most important. Of primary importance is the fact that you can not store two files with the exact same name in the same location (folder/directory) of a hard drive, CD-ROM, or other storage media--without possibly overwriting the other file! Think of the danger that exists if you pull a number of digital image files from various locations to provide to a client, and one file accidentally over-writes another because they are now in the same location. This could also be most confusing, if a client orders an image file by number and you deliver, not the file he is not expecting, but another with the same filename!

The actual file naming system you decide to use may depend on any of the following factors:

These last three limitations are addressed in detail on the filename limitations page.

Beyond ensuring that the name for each image is unique, a filename may help you retrieve an image from its location, and let you know where the physical "original" is to be refiled. In some filing systems it may include additional information about the image that you can tell at a glance. In terms of an image database it has another function, to act as a "hook" to hang additional information, such as meta-data (literally, "data about data") regarding an image. In addition to a description of the image, this might include items such as keywords to find the image, where the image was taken, when it was taken, model or property releases and more.

If you are working with a digital camera, you may wish to sort your images into folders and then automatically rename as a batch function. If you are using Photoshop 7, there's a fairly simple way to "batch-auto-rename" using the file browser. This works similarly with the filebrowser that comes with CS, or Bridge wiht CS2/CS3.

Here's my secret recipe to easily manage your collection of digital or photographic scans (as well as the physical film). The numbering and lettering codes that I use help me to locate the physical film, be it negative, slide, print or Photo-cd. That unique filename can be linked to a database which can store other info about that piece of film. This naming system is application independent, so you can use it with Canto Cumulus, Extensis Portfolio, Filemaker pro, iView MediaPro, Kudo Image Publisher, MS Access, Stockview, or another image management tool of your choice. In addition, I always make it a point to back up the filename as part of the Title field in my IPTC metadata, just in case someone decides to rename the file to make it easier for them to remember the subject in the photo.

The key concepts

Think Virtual Storage First
What I tried to do was create a system that will allow me to give each slide, negative, or digital image a unique name, and keep it within 7 characters. Why not eight? Well, in order to allow more than one version of a file to exist in the same hard drive folder the name must be different by at least one character or number. By using 7 characters to refer to the physical film, the last digit can be reserved to distinguish between variations of a single image.

Create a Subject List For Your Slides
Since slides come in separate cardboard mounts most photographers tend to sort and edit out the bad exposures. Afterwards it's easy to put like subjects together. You probably have some scheme already in place that you may use for your slides. Use that system and build on upon it to cover all the possibilities you can think of. I came up with a list of about twenty primary subject areas, and wrote them out in order. In order to save space in the filename, I broke each down to a two letter code--Agriculture became ag, Animals turned into an, People was shortened to pe. Use your imagination; you probably don't shoot the same subjects I do so my codes may not be of benefit to you. I do write these all as "lower-case" letters to avoid confusing zeros "0" with the letter "O" and though you have to be careful with your lower case letter "l" and the number "1" as they can look very familar.

This same system of subjects will be used to store your digital files. You will have to make judgement calls on some images. An image of a cow in a field could go in either Animal or Agriculture. If the image is primarily of the cow, I'd go with AN, if it's an establishing shot that shows the field, a stream, and the sky as well as a cow, I might decide to put it under AG. Once the image is placed into the image database the meta-data can be used to "cross-reference" the image by describing all the other possible uses for the image, so don't get too hung up on which physical location, or folder on the hard drive that you choose.

Physical Storage Segues to Virtual
After 35mm slides from each subject area mentioned above are assigne a category, they are placed into 20 slide-per-page hanging files. Each page is given the 2 letter code and a 3 digit sequential page number (001, 002, etc). Each position on the page (1 to 20 for 35mm) can be given a number as well. Thus each slide has a unique name/number and each name/number tells me where it is located. The first slide in the AG page would be AG00101. PE10020, would be the slide on page 100 of people section in the 20th position.

Everybody has their own way for storing negatives. For about the last ten years I've been storing each sheet of sleeved negatives, sequentially. I had started out with 4 digits, but expanded that to five by adding a 0 (zero) in front of all the old numbers. That means when I get to 99999 I'm either going to have my own personal Y2K crisis, or I'll have to retire! Other photographers have used the year, month, day (YYMMDD), combination of year and sequential number (YY001 or YYYY001), or other name and number combinations.

Usually when photographers refer to contact sheets we will use frame numbers, but in order to avoid "a" frames (where the image falls halfway between two numbers), I opted, instead, to use letters. The first letter represents the row, and the second the position within that row. So the first frame of the first row could be written as "aa" and the fifth frame of the 4th row would be "de." The added advantage is that mixing the letters and numbers makes the filename a little easier to remember than a string of seven numbers.

This is also a benefit if you are working with older medium format or large format film, as some don't contain any frame numbers. I shoot a lot of bulk rolled B&W infra-red film and it doesn't have any edge numbering at all. With this system it doesn't matter, as I'm going by the image position on the contact sheet to ensure that each image has a unique name. If you shoot medium or large format transparencies and store them in plastic sleeves this may be a better option for storing and numbering. If you have historical prints with no original negative you may wish to use something similar to the method described above for 35mm slides.

The "Mystery" character
If you've been following so far you'll see that I've only used seven characters or numbers to describe each image. However, in the beginning I mentioned that the system was set up to subscribe to the legacy 8.3 DOS filenaming convention. This is for two reasons. The first is that most people have trouble remembering more than 7 numbers or characters. You've probably heard of the "seven plus or minus two" rule. It dictates the amount of information that the average person can store in their short term memory according to a 1956 study (alternate source for 1956 study which includes German and Russian translations) by psychologist, George Miller.(note that there are number of articles which refute Miller's study as well, but in practical use I think it's important to note that most people will have trouble remembering filenames as they are made longer). The second reason is because that last "mystery" character is used to distinguish each version of the digital scans. I refer to this as the image "suffix" and the code for each of those I use is below..

My "suffix" code:

a= Access (low res file for FPO use) If from photo cd then it's the same as Photo-cd Base resolution- 512 x 768 pixels, if from sprintscan or other scanner usually saved as a 6 x 9 to 8 x 12 inch image at 72ppi. (since it's the first letter of the alphabet it will show up first when you do a search in an image database or with filefind or sherlock).
b = Black & white (any file in grayscale color space)
c = CMYK (any file in this color space)
d = Digital
f = Photo-cd scan (Full resolution, Base*16 resolution, 2048 x 3072 pixels)
h = Photo-cd scan (High resolution, Base*4 resolution, 1024 x 1536 pixels)
l = Lab or CIElab color (an archive file in the widest possible color space)
p = Photo-cd scan (Pro resolution, Base*64 resolution, 4096 x 6144 pixels)
r = RGB (generally an archive image or largest possible file size for that format)
s = sRGB (often an access file converted to sRGB for use in PowerPoint, or other applications that are not color management savvy)
w= World wide web usage (images that are web ready, stored as Grayscale or RGB.JPG)

Feel free to create your own system for the types of files you use if these don't work for you.

I use standard file extensions on all image files as if they get e-mailed, uploaded to a web page or burned to cd-rom they will already be cross-platform compatible..PSD for photoshop layered files, .TIF for TIFF files, .EPS for Encapsulated PostScript, .JPG for JPEG images, and .GIF for 256 color, Graphic Interchange Format.

The Form has a Function
So each scanned image follows the form [UID] [suffix].[extension] and thus, all subscribe to the older windows 8.3 file format standard. You can burn all your images to a CD ROM using the ISO-9660 level 1 standard and it will be readable on Macs, PCs, Unix, and even amiga machines.

Example: an02002[suffix].[ext]
an = Animal section, 020 = page #, 02 = position #

click to enlarge

Example: 12035gd[suffix].[ext]
Negative filing number(five digits), followed by two letters representing row and frame. In a roll of film cut to seven strips of five frames, there would be 7 rows (A to G) and five frames (A to E) so first frame would be aa and last would be gd. Six rows of six, would be aa to ff. Using letters avoids confusion in determining the file number from page position, and only requires two spaces (often the center of the frame ends up as 1a, rather than 1).

click to enlarge

Photo-cd files
Example: 0340-01[suffix].[ext]

As I also have a sizeable number of images already scanned in the Kodak Master and Pro Photo-cd format, I've added another type of filename. For this I use a modified version of the disc number and frame number to identify the source of the file.

The first 4 numbers are the last four digits from the Photo-cd jewel case (in large bold type, here it's disc #0340) the second two represent the image or frame number as printed on the index print.

As these are from negs or slides, the actual location of the original can be noted in the "Title" field of the IPTC/File Info section of the image. Which, did I mention before, most image databases will gather this IPTC metadata automatically?

click to enlarge

Digital Camera files
Example: pe10500[suffix].[ext]
pe = People section, 10500 = the 10,500th image in the People folder on your hard drive.

Keeping the letters all "lower case" makes it harder to confuse an "o" with a "0" (zero).

If you need to apply your filenaming scheme to a number of existing images, you can do so using the "batch-auto-rename" feature of the file browser in Photoshop 7.

Once you come up with a system that works for you then you are ready to prepare your images for meta-data encoding. This might include putting appropriate keywords and descriptions into your TIF or Jpeg files using the File Info feature within Photoshop, or an external program such as the Image Info Toolkit, or Photo Mechanic. In addition you may want to read how you can automatically have your filename written into the Title field of your IPTC metadata, as a way of identifying images in case someone changes the filename.

For another perspective, read Ernest H. Robl's article, Image Numbering, Filing, and Retrieval that was originally prepared for the American Society of Picture Professionals (ASPP) site.


examples  |  books  |  products  |  image databases  |  links  |  what's new
imagedatabases  |  programs  |  IPTC standard  |  downsampling  | filenaming 
metalogging  |  captioning  |  keywording  |  guidelines  | metalog resources
home  |  contact  | sitemap