Image File Naming

In my last post (Scan, Scan, Scan), I wrote about how I got so far into scanning my photos. With (now) over 135,000 photos in my Lightroom catalog, and more than 20,000 of them scanned from film (I just did a more accurate count - I'm close to 27,000 scanned photos), I had to come up with a naming strategy that works long term.

Early in my photography days, I wrote descriptions on the frames of my 2x2-mounted 35mm slides. Something like "Playing Baseball in Front Yard". When I first started scanning, that was my scanned image name. It didn't take long to figure out that this wouldn't work. What if I had two photos with the same subject? So I started adding "-1" or "-2" to the names. It was clear this wouldn't work for long, either.

I tried a number of different schemes, but didn't hit on anything I really liked. That is, until I ran across a book, The DAM Book, Digital Asset Management for Photographers, that gave me the insight I needed. The Second Edition of this book is worth having on your bookshelf if you take a lot of photos. (There is way more than file naming information in this book.)

My filenames now look like this: gbs_20170304_5156.dng. Let me expand on this for you:

• gbs - my initials. If I scan photos belonging to someone else, I use their initials. 
• 20170304 - The date. That should be obvious. YYYYMMDD - 4 digits for the year, 2 digits for the month, and two digits for the day. Month and day are always 2 digits. For example February 15, 1995 would be 19950215. This way you can sort chronologically just by ordering the files by filename. 
• 5156 - the sequence number from my digital camera. If the camera uses a different numbering scheme, then I use that. For instance, files from my Panasonic Point-and-Shoot camera end up looking like this: gbs_20170219_1040995.jpg. And from my Android phone, like this: gbs_2017-02-17 14.32.55.jpg (OK, that sort of breaks the sorting rules, but that's the way it is).
• And, obviously, the file type - .dng (Digital Negative), .jpg (JPEG), etc. 

The good thing is that Lightroom lets me import my photos from my camera into this (these) format(s) very easily, so there is no manual renaming needed. 

What about scanned photos? They don't have an embedded sequence number. Again, I've done lots of experimenting through the years. The bottom line is that I only need to 1) uniquely name the files, and 2) be able to sort them into some semblance of order. 

I started out naming files like this (before I had completely formed my naming standard): gbs_19950721(05). The number in parentheses at the end was the frame number from the roll of film. That worked... until I shot two rolls of film in a day. Then I added 100 to the frame number for the next roll, becoming gbs_19950721(105). But that felt clumsy. And did I reset the number for the next date, or did I keep the 100 until I finished the roll?

Since ancient times (1960s or 1970s), I had kept my rolls of film numbered through the year, so a roll of film would be Roll (for example) 1988-12. Duh! Simple! I don't think I ever shot more than 100 rolls of film in a year, so I started numbering my images like this: gbs_19880325_1206. The 1206 is frame 6 of roll 12. And if you shot more than I did, just make the roll number 3 digits long. 

That leaves only one set of images with which that I don't follow this standard. I have a few boxes of loose black and white negatives. One of those boxes appears to be photos my grandfather shot starting around 1920 and continuing at least into the 1940s. There are no dates. There are no frame numbers. The rules are completely broken. At this time I'm just trying to preserve them. I'm placing them in folders that indicate who I think shot them, and they are simply named Image001.jpg, Image002.jpg, etc. As I find out more about them, I rename and catalog them. For instance, I found a photo of my Mother's first birthday - I can come pretty close to when that photo was shot , so I've renamed it accordingly. Of course, the sequence number is fiction.

Stay tuned. My on-disk organization might be next (or not). 

Of course, I must include a photo. This is photo gbs_19740317_0012.tif. Clearly it was shot March 17, 1974. It is the 12th frame on the roll of film, and I scanned this photo before I fully developed my naming standard. The photo shows Kettle Falls on the Columbia River near the town of Kettle Falls, Washington. It was taken the last time Kettle Falls was visible when the level of Lake Roosevelt was lowered during construction of the Third Powerhouse of Grand Coulee Dam. 

Scan, Scan, Scan

It all started simply enough. I had an HP flatbed scanner (I don't recall the model) that came with this little mirror device that would redirect light from the scanner back down through a 2"x2" mounted 35mm slide so that you could scan the slide. Resolution? Oh, probably 400 or 600 dpi. It worked OK (just OK) for well-exposed or slightly over-exposed slides. Not so well for underexposed slides. I may have scanned a few dozen slides, but the quality wasn't adequate for me.

Some time later, I purchased a dedicated film scanner (Nikon Coolscan 3000).

The Nikon Coolscan 3000 could scan up to 2,700 dpi. At first, I thought this was more resolution than I needed, so scanned many slides and negatives at 1,350 dpi. About enough for a decent 4"x6" print.

It wasn't long before I wanted better. So I purchased a Nikon Super Coolscan 4000 ED film scanner, capable of 4,000 dpi resolution. And I upped my default scanning resolution to a whopping 2,000 dpi, scanning some specific negatives and slides at 3,000 or even 4,000 dpi if I thought I wanted larger prints from them. Eventually, disk space got really cheap and I realized that I should just scan everything at the maximum resolution. So all 35mm film I've scanned for the last 4 years or so is at 4,000 dpi. Eventually I purchased a Nikon Super Coolscan 5000 ED film scanner, which also scans at 4,000 dpi, but had a few more features that hopefully would improve my scan quality.

For image viewing and printing software, I used ACDSee, and later, ACDSee Pro. Not sure when I purchased this, but know I owned it in 2001. I still use it for specific purposes, which I will describe in a future blog.

When my friend/neighbor Jenne was about 7th grade (around 2008), I got this idea to create a photo book for her for her senior graduation (plan ahead...). I had been shooting all digital since 2003 or 2004, but had known Jenne since her arrival and adoption in the United States in 1998, so had around 6 years of photos on film. Before this time I had occasionally scanned some of my photos, but hadn't gotten really serious about it. I had probably scanned a thousand images.

So I started scanning photos of her. But, where to stop? What if the roll of film I was scanning had some other interesting photos? Should I not scan those? Or scan those? I opted to scan all. My scanner/software could scan a strip of negatives at a time, so it was easy to insert a strip and let it automatically do a preview of all negatives in the strip (usually 4), then select which to scan. But I also wanted to keep track of which rolls of film and which negatives had been scanned. The scanner software could accept a filename pattern, and automatically increment a number within that pattern. So I decided to make the filenames include the negative number.

To start with (I think - this was years ago, and I've been refining my scheme through the years, and don't recall all details, or the exact order of the evolution), I put the images from each roll of film in its own folder, and labeled them something like IMAGE00l.JPG, IMAGE002.JPG, etc. More about the evolution of my file naming conventions in a future blog.

Now that I had a goal, and a deadline (Jenne's graduation), I started scanning everything I had shot since 1998 up until I was shooting all digital (around 2005). I had tried multiple tools for cataloging through the years, but they each had its own proprietary database. At one point, an upgrade to a new version of a tool lost all of the data from the previous version. I knew what I wanted, and it wasn't available. I looked at a professional digital asset management tool that was, in my mind, too complex (and quite expensive). I knew the only way this could work was to store the cataloging information in the image file.

And I continued scanning.

To go back a few years... Cataloging my photos was something I have thought about since early in my photography days. I started shooting when I was around 10 or 11 when my dad gave me his old Kodak 620 folding camera (I still have it). I then upgraded (?) to a Kodak Instamatic with a little pop-up flash that used 126 film. When I graduated high school, I could still look at any photo I had shot, and recall the circumstances of the photograph. By the time I got through college, I could still remember pretty much any photo, but storage was becoming an issue. I was shooting all 35mm slides at that point, and was filing them by subject matter in slide storage boxes. Being able to remember any shot allowed this to work. If I wanted a particular photo, I could remember where I had filed it.

I had subjects such as Family, Friends, Church, Scenery... I knew this wasn't going to work much longer. How should I file a slide that had pictures of, for example, Family at Church? So I switched to filing slides chronologically. Sometime in the mid-'70s I learned about Key Word In Context (KWIC) (https://en.wikipedia.org/wiki/Key_Word_in_Context). I wrote a computer program to implement this. Here's a simple example of how it worked.

Say I had a photo (labeled Photo 045) that I described as "Dad and Randy at Mt. Rainier." The KWIC program would ignore a list of connector and filler words (such as "at" and "Mt."), and generate one line of output for each significant work in the sentence. Thus this description would generate 3 lines of output:

1. Dad and Randy at Mt. Rainier - (Photo 045)
2. Randy at Mt. Rainier - Dad and (Photo 045)
3. Rainier - Dad and Randy at Mt. (Photo 045)

Let's say I also had Photo 060 with a description of "Mom and Dad at Christmas." This description also generates three lines:

1. Mom and Dad at Christmas - (Photo 060)
2. Dad at Christmas - Mom and (Photo 060)
3. Christmas - Mom and Dad at (Photo 060)

Now I would have an output file that could have thousands of lines of output for maybe 1,000 photo descriptions. It would then be sorted alphabetically, yielding:

1. Christmas - Mom and Dad at (Photo 060)
2. Dad and Randy at Mt. Rainier - (Photo 045)
3. Dad at Christmas - Mom and (Photo 060)
4. Mom and Dad at Christmas - (Photo 060)
5. Rainier - Dad and Randy at Mt. (Photo 045)
6. Randy at Mt. Rainier - Dad and (Photo 045)

and I could now scan down the list for the subjects of interest and locate the photos. In reality, after I had entered about 100 photos, I realized this was quickly going to become too cumbersome, so I abandoned it.

In 2006, Adobe announced the beta of Lightroom. It could catalog. It could edit photos. And it stored the data in the image file! I immediately downloaded it, and started learning how to use it. (I'm still learning.) I started tagging images with keywords, and those are all stored in a database and in the image file. My dream had become reality.

By then I had thousands of images on disk, as I had started shooting digital in 2002, plus I had scanned a "few" slides and negatives. I tried to go back and catalog everything, but I missed many hundreds of photos, and my cataloging was incomplete. But at least I could quickly and easily locate all images I had cataloged.

Now I could continue scanning. And cataloging both the scanned images and the ones shot on digital cameras. And continue scanning I am doing. I've scanned photos for friends. I've scanned negatives my dad shot in the Army, and of me when I was a year or two old. And negatives my Grandpa shot when my Mom was 1 year old, and through her high school years, and of me. I have 100-200 rolls of negatives that I got from my Mom that she has shot before switching to digital that I am scanning. And cataloging. She also has several (at least) file boxes of slides, of which I have only scanned one or two.

All told, I now have, at this moment, 134,620 photos in Lightroom. I have scanned somewhere over 20,000 of those, and the rest are from digital cameras, mine, family and friends.

And, yes, I completed the photo book for Jenne. It consisted of about 800 photos, beginning with her arrival in the United States of America through her high school graduation. I was a little late to give it as a graduation present, but it did make a good Christmas present that year. In case you are wondering, I have 10,247 photos cataloged of Jenne. I'm sure I've missed a few. And, over the years, Jenne helped me pick the photos she liked. She didn't know the goal, but I would occasionally show here several hundred photos that I had selected, and have her mark (with a special key word) the ones she liked. Thank you for your help, Jenne!

And her younger sister is now only four years from graduation. I have another task ahead of me. The good news is the film photos of her are already scanned.

Scanning these photos allows me to share these old photos with friends and family. And it brings back good memories. Here is a photo of the sails on the ship Fantome on a cruise in the Caribbean in 1991.