On image database keys

Christmas Punks, #1

Christmas Punks, #1

Key: R20101225-095142-toned

I got an email from a new Flickr viewer, Joseph, who asks:

Greetings,

I got intrigued by your KEY system, such as (example above).

Could you please tell me a bit about it, or point me to an explanation of it?

Thank you,

Joseph

Since other readers may perhaps be interested, let me answer here.

The key that appears next to each image is a key that I assign to uniquely identify the image. I chose this key system because image databases come and go every few years with the latest hot product. I wanted a system that I could use for a lifetime–one in which I am in control of choosing the key. When I post an image either to my blog, Flickr, etc., I like to put the key next to it. I also usually put the key on my prints as well, either somewhere unobtrusive on the mat or on the back. That way if anybody is referring to an image of mine in a conversation with me, we can jointly identify it absolutely and quickly without resorting to descriptive terms, which may match many, many pictures of mine with similar content or themes. Such conversations come up all the time, whether someone is asking how I did such and such technique, or where did I take that picture, or who is it in that picture, or can I get a copy of that picture, etc. etc.

The key system I chose is based on time, the eternal dimension which governs everything I do and provides a natural counter. The form of the key is naturally derived from the EXIF information recorded in the image metadata. When I read new images off of my cards, no matter which kind of camera, I run a Python program (of my own design) which reads the EXIF information and renames the image files according to the time the photo was taken. The format is

YYYYMMDD-HHMMSS[-NNN][-tag][...][.ext]


Typically I am not taking more than one image a second, so the “base” frame will end in HHMMSS. If I shot a multi frame burst, I might end up with suffixes of 001, 002, etc. for the “NNN” part. Finally, if I edit the image in post, and make a derivative image, I will often add some brief tag to the end to tell me what main adjustments I might have made (e.g. -crop, -curves, etc). If the image was heavily edited, I might just append -master.

Since the image filename is named after the key, I can store the file anywhere and with only a glance can tell exactly when the image was taken and most likely what major tweaks I did in post. I tend to store general images in folders by year, except for special trips or projects, which have their own folders. With this organization, I can find the original file, master file or other derivative files quickly given the key.

For image metadata (basically keywording, to find images having to do with certain themes), I just don’t do that much. I do create occasional XML files that contain lists of image keys and some associated metadata with them. Frankly, though, for my needs as a photographer pursuing only my own art and passion, I don’t have much need for that and my time is too valuable now to be spending lots of time adding metadata to my images.

So there you have it. Works for me, YMMV!

Christmas Punks, #2

Christmas Punks, #2

Key: R20101225-095055-toned

2 thoughts on “On image database keys

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: