Renaming pre-2000 files

[David Arnold]

Do 2 extra digits really matter that much?

At times I'll be adding characters at the end to identify derivatives (per The DAM Book, pages 81-82), and as the number of characters gets into the high 20s (which would occur with derivative files), when the thumbnails are displayed in iView at my preferred size the entire filename isn't visable.

If you're worried about length why use your full name and not your initials or another shortening?[/i]

Reasonable question. The answer is, in a word, branding. As Peter points out (page 79), there are benefits to making sure people know who an image came from. He does suggest initials if names are long, but that isn't as identifiable. I (last name Arnold) and my wife (last name Rutman) are stock photographers and freelance photojournalists. We work together closely, all photos are copyrighted in both names, and published photos credit both of us. I tried filenames starting with DAGR_ or DA&GR_, but neither would accompllish the same thing as ArnoldRutman_

It's not yet locked in stone (or I wouldn't bother posting the issue here), so I'm still open to any better ideas or arguments.

David

[DJ Webb]

I see where your coming from David and -99 _00 is a good work around to the sorting problems. Its not like your gonna get to 2099 (or earlier) with this system.

I currently use my initials in my filenaming schema (DJW) which i think will be a good enough indicator to me if i see a photo i think is mine somewhere else. I agree that using your "trade" name is beneficial especially in your field.

[David Arnold]

Having finally settled on a naming system I'll stick with (after backtracking from a couple I tried and gave up on over the last few months), I thought I should report on my journey.

(1) Tried using underscores (_) and dashes (-) for different purposes, dropped that fast. Much to easy to misread.

(2) Tried YYMMDD_ for the date field, dropped it. But not for the issue I was concerned with (images from the '70s, '80s, and '90s coming before newer ones). Being so used to automatically reading a 6-digit date field as MMDDYY, I was constantly misreading dates on individual files, and mis-typing dates on folders.

(3) Tried ArnoldRutman_ for the first part of the filename, stayed with it for a while, but when I needed to change the date field from 6 characters to 8, I decided the whole thing was getting too long, and went to AR_.

(4) After lengthening the date field, I looked at our shooting (you'll recall my wife and I shoot together, and after initial edit merge our files; all images copyrighted in both names), I figured we were unlikely to ever need more than a 3-digit unique field (we've shot more than 1,000 images in a day, but do an initial edit before filenaming, and can't imaging having more than 1,000 keepers for a single day's shooting), so for a few months we used a 3-digit field. However as I continue to gain experience with Peter's system, reread his book, and follow the threads on this site, I keep finding myself dropping my bright ideas and going back his way of doing things. While the early culling works for us now, I can imagine the possibility of switching at some point and making numbering the first step. Iin that case a 3-digit unique number field would leave me (depending on your preference for wet or dry metaphors) in deep eoo-doo or up the creek without a paddle.

So....we've finally settled on AR_YYYYMMDD_####.ext (e.g., AR_20060908_0083.CR2).

[ralphpaonessa]

I also sell stock and did not want to highlight the shot date for fear the buyers would be prejudiced against older images. I also want to integrate my pre-2000 images into the library. My quirky solution was to start the file name with "1" for year 19xx, and "2" for year 2xxx. So shots from 2006 begin "206-" and shots from 1996 begin "196-"

The actual dates are included in the metadata, and an observant person could quickly decode that 206-10-1234 was shot in 2006. But that's my approach. And it makes the filename one digit shorter!

I do want as short a filename as possible, because at some point a buyer will tell/write me that he/she wants 203-24-0031, which seems more manageable to transcribe than Paonessa_20030821-163459-45376.

I have been struggling with a naming system. Besides 206- for this year, I manually add a 2-digit code that roughly signifies the "shoot" I did, followed by 4 digits in sequence for up to 10,000 images in that shoot. I tend to do a few major trips, each with many images, so I'm not in any real danger of running of out 2-digit codes. And it's just for added convenience that I group shoots that way. If I started doing hundreds of small shoots, I could put many shoots under 206-25-xxxx.

My requirements for a naming/numbering system are:
1. Unique filename for every image
2. Files are named (numbered) in order shot.
3. Keep name as short as possible for easy transcription.
4. Avoid letters, which are easy to confuse when reading or speaking (6 vs. b, 1 vs. i. vs l, C-B-D-E-G-P-T-V-Z all sound similar)

I would love to have a system that automatically renames/numbers images as they are downloaded from the card. But that requires the following temporary names:

YYYYMMDD-HHMMSS-nnnn

where nnnn is the last 4 digits of the camera filename. (I shoot with an 8 fps motordrive, as well as multiple cameras, and my Canon bodies don't record sub-second timings, so I add -nnnn from the original filename to distinguish all the images shot in the same second.)

But this 18-digit number can accommodate one billion billion images, and so it's a very "long-winded" way to number my eventual hundreds of thousands (millions?) of images.

So my solution (which I'm about to implement, but open to suggestions) is to rename these images sequentially into my 2YY-xx-0001 system. I do this after all the images from a trip are downloaded and combined (including from multiple cameras) and sorted in the order shot.

The initial downloading and renaming will be done (I think) by Downloader Pro from Breeze Systems, which has a flexible renaming capability. The final renaming can be done by Bridge or iView.

My final comment is that my only absolute requirement is that each and every original image have a unique filename, and a unique place it's stored, recorded in my database. (I do this with my slides also -- remember that old analog technology?) As long as I have a good database, all the other information tha might be encoded into the name (making it longer) is available in my database.

Now, if iView goes south on me, I'm going to be a VERY unhappy camper!

Ralph