Field projects generate a huge number of photos. Those photos are also used at every phase of the project: writing field reports at the end of the project, research during the off-season, presentation of preliminary results in lectures, illustration for publication, and finally repository in (hopefully) some useable searchable form for others to use.

Too often projects rely on their own folder techniques. That is, someone creates a series of folders and sub-folders to hold the images and that folder follows them wherever they go. There are obvious problems with this approach. The most unfortunate is that the images never get cross-indexed. Whether you file things by date or subject, you often end up wanting a particular photograph for a different context. If you file all photos by trench and photo date, how can you find all photos of water pipes, for example? The second most obvious issue is that, if the photos are organized by subject, who does the organizing? I have worked on projects where the trench supervisors organize their own trench photos and then hand them in at the end of the season. The result is a series of folders with idiosyncratic organizational schemes that promises time wasted looking for a particular photo.

I have worked a long time to get a technique that fixes all of these problems, including the archival issue, and while it might look complex at first, once set up it is fairly painless. And it allows you to search for an image that you want without a database.

The technique is Mac specific and requires two additional pieces of software: Graphic Converter and a plug-in for FileMaker named Troi File as well as an Automator script. You only need one copy of Troi File, since it is only used when the data is entered and not when browsing or searching the data.

The overview of the method is as follows:

  1. Copy the raw images from the camera to a particular folder on the hard drive.
  2. Use Graphic Converter (Mac) to review the files and rename them according to the PARP:PS digital numbering scheme.
  3. Run an Automator script on the images. This will make copies of several sizes of the images for archiving, importing, and web serving. The Automator script will also handle the next step:
  4. Import the images into the database. This is also scripted so the database also imports the two sizes of thumbnails as well as the metadata from the images and parses it out to certain fields. Handling the metadata of the image requires a plug-in for Filemaker and will not work from all computers, so make sure that the first four steps are all done on the one main computer.
  5. In the Media List layout of the database (on any computer) enter a caption and any necessary numbers that will link the subject of the photos to their records in the other data tables.
  6. Export any caption data from the database and insert this into the IPTC data field of the archival and web-sized versions of the images.

What you get when you are finished is a folder with the original full-sized images as well as two additional copies of the images in smaller sizes all in a single folder on the desktop. One size is used with the computer’s built-in web server, the other is used for importing a rather large thumbnail into the database. We found that the default thumbnail size was too small for most uses. In the database you will have a record for each image with the camera metadata in fields in the database (we are mostly concerned with the date and time the image was taken). You should have descriptions and identification numbers of everything in the photo and links to a PowerPoint sized version of the photo suitable for use in a field report or presentation.

After the caption data is added to the database we then export that data as a text file and insert it into the image using Graphic Converter. This solves the problem that most people have with trusting digital images, especially if you don’t have the subject metadata in the image itself. Years ago it was assumed that a proper archival procedure would be to print the contents of the images database and archive that along with the digital photos in case the original database file format is lost. But the IPTC data does that for us in the image itself. It also allows us to search all of the images without the database. If properly filled out, I can search for an SU on any image saved to my own computer. All of the images that have been identified as having that SU will show up in the search results.

The entire kit can be downloaded from here. This contains the database (again, but I made some changes detailed in the ReadMe), the automator script in both application format and editable workflow format, and more detailed step-by-step instructions in both rtfd and html format (again so they can be edited).