Pushing Pixels

This weekend I sat down for a periodic cleanup of my hard drives. This is one of those things that always gets away from you, and the longer you let it go, the tougher it gets to put back in order. Basically, if you are not careful to put files back were they belong, it can become a huge headache later when your trying to "find that one sweet shot from a few years ago"

Some common problems:

Different versions of the same file in different locations The same file with multiple names Live versions with broken links to referenced files.

Now, I was planning on doing a screencast of my work flow and an accompanying post on how I store and back up files, but that is probably far in the future (and Chase Jarvis just did it). Instead, I will try to lay it out for you short hand. If you get lost while I am talking about this stuff, then that is the first sign that you need to read up on proper workflow. When I was new to the game, a lot of this stuff was pretty confusing and I really didn't understand the reasoning behind some of it. After using the following method for several years now, I can say that it will save you a lot of time and frustration in the long run, and if you shoot photos for a living it will eventually save your butt when something goes wrong (and it will, eventually). I would say that, after a camera, lens, card, computer, etc. your next purchase should be a matched pair of hard drives to store your precious work on. No, hard drives are not as sexy as that new ttl flash that you've been eying up, but when the day comes that your computer crashes, or is smashed to bits, or your roommate spills his drink on it, you will be glad that your data is safe. Below are the steps that I use to manage my photos and video.

1) Save exact copies of all of your files in at least two places. There is really no excuse for not backing up your work. Hard drives are really cheap these days, and even the free software that comes with your camera will have a dialog for saving files in multiple locations right from your camera. When you watch the video you will see almost the exact method that I use for importing image files off cards. "Location one" is the first drive and "location 2" is the second. Later in the video, Chase goes even further, and moves the files both onto the server (very secure), and another raid 1 array off site. This essentially means that the whole server would have to be destroyed (unlikely) to lose the data there, and even if it did there are two additional copies of the file that are safe in another location.

2)The second most important thing is that your files are consistently labeled in a way that is meaningful years down the road. This way, you will be able to find a specific image or file when you need it. The files the come off my Nikon dSLR look something like this: DCS_9456. This would initially work for cataloging files, but there is a major flaw with using this naming system; the counter on my camera resets every 10000 images. As it happens, the above example is the 49456th image that I took with this camera, but because the camera counter resets every 10k there is no way of knowing that without looking at the metadata. This also means that if I had used that defult naming system, there would be 5 different images in my system that have the same name, not ideal. Instead I use this system: 20101101_picnic_0001. The file name begins with the reverse date of the day it was shot, in this case YEAR: 2010, MONTH: November, DAY: 1st. After that, there is a tag that describes what is in the folder, this is generally a single word, just enough to identify the subject of the project. Finally there is an ID number given to each image, in this example the ID number is 0001. This particular photo would belong to a folder named 20101101_picnic. By labeling both the image files and the folders this way, each shoot will be arranged on the drive in the order it was saved and it will be easy to find a particular image down the road with some basic info like the date and project name.

3) The third thing to realize is that this is just how a store the raw image files. I will also create a Aperture library for each project and label it the same way. Aperture is a non-destructive raw editor. When you open a file in aperture, you are being shown a preview of the raw file on your hard drive. As you edit the photo it makes a tiny rider file that keeps track of all of the changes that you make, without actually changing the raw file. This information is stored in the Aperture library. When you export an edited file from Aperture, the program saves a copy of raw file with the changes you made to a separate file that you save in a different. Chase calls this "Live Work". I think that this is a good way of doing it. If you actually made changes to a raw file, you would have to update your backup of that file as well. I have about 3.5 TB of raw files, and if you had to update all that data every time you made a change to 1 file, you would go crazy pretty quick. It would take like 8 hours to do the update every time, the constant overwriting would also increase the chance of corrupting your precious data and wear out your drives pretty quickly. To solve this problem, I also use a "Live Work" folder, saved in a separate area of my system. I only have about 10 GB of data in my live work. This is because even though I might shoot 1000 photos in a day, I am only choosing the best 2 or 3 to work on and deliver to a client. Even on a large project, a may only choose 1% 0r 2% of the total images to deliver. I pretty much follow Chase on this part of the process, so watch the video for a little more detail.

That's it! That is really all you need to do to protect your photos and other files. By saving your data this way, you are creating a situation where there would have to be many simultaneous failures before you actually risk losing anything. Once you are in the habit of working this way, it is very easy to keep track of a huge about of data, and the whole process really doesn't take much more time then dragging the files off of you camera and into a random folder on your desktop. It is important to note that this is just what works best for me, there are many ways to save and backup photos. Take a minute to watch this awesome video by Chase Jarvis, he can explain this stuff with a little more detail.

(Updated) For the record, I use a pair of G-Tech 500gb G-Drive Mini's travel with my laptop. When I get home I dump them onto a my server, which is a Dell Dimension E521 running Linux, that drives a pair of rack mounted 8tb boxes. I used the same system for my design work.