My Approach to Image Storage and Backups

In a digital age, and especially with photography being so prolific these days, an important issue for photographers to consider is their approach to file storage and backups.

I have been photographing digitally since 2002, and in my early days I found a method of filesystem organisation which works well for me.

My backup strategy also works well for me, and as I’ll explain further into this article, provides me with disk-level redundancy as well as geographical redundancy.

1.  Filesystem Organisation

At the time of writing, I have 32,969 image files (ie, with a filename extension of .bmp, .cr2, .gif, .jpg, .png, .psd and .tif).  Another 1,075 files are .xmp files, which contain metadata written by Adobe Camera Raw during the raw conversion phase.

That’s a lot of files!

It would be impossible to manage without a good directory structure.

The system upon which I decided in the early days was to have one directory for each year, and within a given year, a directory for each shoot, named after the shoot.  Also a part of the directory name is the date of the shoot, in YYYY-MM-DD format.

Here is a slightly dated screen capture of my directory structure:

My Image Directory Structure

My Image Directory Structure

As can be seen, each shoot has its own aptly-named directory within the year in which the shoot took place.

Now, some people place the date first, but my mind is wired to remember the location or event rather than the date it took place.  To my mind, “I shot at Long Reef at some stage this year” holds more meaning than “I shot somewhere on the 21st of March”.

Other people break down their directory structure further, with a subdirectory for each month.  Again, the specific date on which I did a photoshoot doesn’t rank anywhere near as important as what the subject was.  Being able to pinpoint a shoot to a specific year by memory is as low-level I need to get with dates.

As for image filenames, I don’t rename the files the camera produces.

If I shoot an image and the camera calls it IMG_0001.CR2, and I decide to process it, I will end up with at least three more files:

  1. IMG_0001.XMP (the metadata containing the non-destructive adjustments applied in Adobe Camera Raw);
  2. IMG_0001 Processed.PSD (the 16-bit, multi-layered, Photoshop-processed version of my image); and
  3. IMG_0001 Processed.JPG (the 8-bit JPG exported from the PSD file).

I may produce smaller versions of my images for specific purposes, or black and white versions, and in those cases the exported JPGs would be called “IMG_0001 Processed 1024×683.JPG” or “IMG_0001 Processed Black and White.JPG” respectively.

I don’t keep any lower-level directories for processed versions, ‘keepers or any other category; all files from a single shoot are in a single directory.

Naturally, the image filename counter on a camera will reset once it reaches a certain frame count, and as it stands, I have three IMG_0001 files in some cases, but filename duplication is not a cause for concern, as a directory containing a single photoshoot’s images will never have two identically-named files.  On the few occasions when I have used two cameras to cover an event (such as a band), I have kept each camera’s images in its own subdirectory.

One organisational strategy I don’t employ is image tagging.  My existing filing strategy has been in use for quite a number of years, and I haven’t found a need to search for an image’s contents based on something I remember being in the image, or any other criteria for which tagging  would be useful.  I’m able to find what I’m after by recalling the year and the event/shoot, based on the naming convention I use.

2.  Backup Strategy

I take backing up quite seriously, and I have adopted an approach that provides disk-based and geographical redundancy.

My PC is a Apple MacBook Pro (17″) with a 500GB hard disk.  All of my image (and other) data is stored on it.  In my home office I keep two LaCie d2 Quadra 1TB external hard disks, which have a FireWire 800 port offering faster data transfer rates.

I keep a third external hard disk off-site to ensure geographical redundancy in case my local external disks suffer loss, theft or damage.  My off-site disk is a 500GB Western Digital My Passport Essential.

My approach is to manually back up my data to these drives, generally every week or two.  This method might sound cumbersome, but it works for me, and I employ a very structured filing system, not only for my image data as outlined above, but for all of my data.

At some stage I should investigate a more automated method, but this manual approach has worked for a number of years.

The end result is that I have four copies of all of my data, some of which goes back 16 years.   (The current value of that 16-year-old data is questionable, but I am sentimental.)

In the past I used CD media, and later DVD media, but as my volume of data increased, and as image file sizes and the byte count of individual photoshoots also increased, optical media quickly became cumbersome and insufficient, and that was when I invested in external hard disks.

The other issue is that I don’t trust optical media.  Sure, hard disks can fail, too, but the key is not to rely on one disk.  Hard disks are much more convenient to use, a lot faster, and they store a lot more data.

To conclude, my key advice on data backups is:

  1. back up regularly (weekly or fortnightly is good);
  2. back up to multiple storage devices, never relying on one alone;
  3. keep your data in at least two geographically separate locations; and
  4. use the most effective and reliable storage media of the day, and update when needed.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s