FADGI Struggles

We're going to get a little technical with this post and talk about digitization standards for archives. The two main goals of archival digitization are access and preservation. Of course, any digitization that works to have digital images of archival materials published online is going to be a win for archival accessibility. However, digital preservation can be a little harder to attain.

This is where the Federal Agencies Digital Guidelines Initiative (FADGI) comes in. FADGI is a "collaborative effort by federal agencies ... to articulate a common sustainable set of technical guidelines, methods, and practices for digitized and born-digital historical, archival, and cultural content" (https://www.digitizationguidelines.gov/about). FADGI is essentially the gold standard for digitization in archives and works off of a 1 to 4 star rating system, with 1-star being the lowest quality and 4-star being the highest quality.

FADGI updates and publishes a guide on several different archival materials every few years. All archives need to do is make sure their digital images match up with the quality standards set forth by FADGI. Except, it's not that simple.

A sample of the FADGI guidelines, simple right?

To check if your digitization equipment is up to par and scanning your material at FADGI standards, you need a target to test with. Below is an example of a color target:

A GoldenThread FADGI testing target from ISA: https://www.imagescienceassociates.com/goldenthread-fadgi-19264-device-target.html

These targets can be purchased through groups affiliated with FADGI and the Library of Congress, like the Image Science Associates (I'm sure you'll notice the hefty price tags on these bad boys). I use a small target for black and white 4x5 negatives. So, once you've made room in your tiny library budget for a target and have taken a scan of it, now what? You have the tested image but don't have anything to test it against. Thankfully there are two current software options: DICE or GoldenThread (you'll need to squeeze more out of your budget for this one!) and OpenDICE (the free but terribly clunky cousin to DICE). I use OpenDICE to test my target scans, but I won't go into the details on how to navigate OpenDICE here (that could be its own tutorial video).

A portion of one of my test results

Here is an example of a portion of one of my test results. I copy my results into a separate spreadsheet and add some color and screenshots of the graphs, otherwise, it would look like a jumble of data. Note how the first section has a bunch of ugly red "Fails"? That's because despite all this information here, I have not managed to successfully achieve even 1-star FADGI compliance. However, I'll keep testing and making adjustments and I'll hopefully reach some sort of FADGI compliance with this project. When I do, I hope to share it here!