lightweight digital archives for archaeology

In the nine years I’ve been in my current position a lot has changed. I started out with a working title of “Archaeology Inventory Manager,” mainly tasked with assigning site numbers and keeping the state’s site forms in order. Other associated paper records and physical media were sorted in alphabetical order by county and stored in various locations in the building over time, depending on space and the primary users of the material. Digital files would sometimes come in on disk with bound reports. We would download them and organize by county. Over time we began to require PDF copies of full reports, but digital photos and datasets have always been sort of pleasant, optional additions to our collection,

This setup worked very well for a long time, but its efficacy relied upon the assumption that researchers (whether internal, commercial, or academic) were physically in the building, browsing through an approachable volume of information. Naming conventions, organization, and discoverability of materials all descend from the way researchers found information within a specific physical space at a given time.

Here we are in 2017. Things are different. The agency’s archaeological site inventory is comprehensive and online. Many researchers don’t even come to our physical collections to do background research, or if they do it’s infrequent and tasks are planned in advance. Much of the information produced by archaeological investigations is born digital, and there is a lot of it. Researchers expect to be able to find information with a few clicks.

The agency faces a lot of technology challenges due to a combination of high security requirements, small IT staff (n = 1, or 0, at any given time), and very high costs for services and network space. Additionally, like most (nearly all?) archaeological repositories, we’re also running out of physical space.

So here are the priorities I’m hoping to address:

  • Storage: find a place to put these files
  • Preservation: make that place a safe one
  • Access: allow researchers to get to it
  • Discoverabilty: easily, by linking records together

Tall order, yes. And there’s no defined budget for any of these facets. So it goes. These are multiple components of several multi-year projects. But first thing’s first.

In the very short term, I’m hoping to set up the bones for a lightweight, low-dependency system of organizing media. Transferring files to hosted (or, “cloud”) storage isn’t an option without many months of red tape negotiation, although I’m investigating some generous offers for help with stable backup. But our working area is still likely going to be on removable storage for a while.

Here’s what I’m thinking for now– Set up directories on removable storage, with BagIt to generate checksums and monitor file integrity. Keep experimenting with TagSpaces Pro for metadata tags. The wonderful thing about this little application is that it creates a sidecar JSON file with user-defined metadata, like this:
tag for a spreadsheet file

Here’s what the JSON sidecar file looks like:

{"tags":[{"title":"plan-view","type":"sidecar","style":"color: #ffffff !important; background-color: #42d692 !important;"}],"description":"","appVersionCreated":"2.8.0","appName" :"tagspaces","appVersionUpdated":"2.8.0","lastUpdated":"2017-06-28T14:50:06.700Z"}

Ideally, when we’re ready to migrate to a more robust or permanent system, we can be ready to go. I’m still experimenting with more utilities I can add to the workflow to make organizing and storing these materials safe and as simple as possible. Stay tuned. This is definitely a grand experiment.

comments powered by Disqus