Two weeks ago, I reported that we reached a significant milestone by scanning the final JFA catalog for the Shared Shelf Project. There’s still much to be done, so this week we’ll look at the work of creating metadata: or the stuff that makes your digital images discoverable.
When I first started the archives project in 2009, a surprising number of people encouraged me to “just digitize it all and post it on-line,” as though it were as easy as waving a magic wand. I knew then and still understand now that digital collections require careful curation and management. Students began scanning especially significant collections in 2010, but without the appropriate infrastructure, these have languished on Google Drive. The only way to access them is to know they are there and request we share them with you.
More recently, these collections have finally begun seeing the light of day. Lacey Wilson ’14 (currently a digital history fellow at the Roy Rozenzweig Center for History and New Media at George Mason University) has created a website, Blueboy Letters, using World War I letters from IC students that she digized while a student at IC. And by the end of the summer, we hope to post Edward Beecher’s unpublished novel, Cornelia, on Shared Shelf as well.
Overall, I’m quite happy with our progress on the digital front. In September 2015, one of the central messages at the first national workshop for the Consortium on Digital Resources for Teaching and Research, was “Be glad you are starting this now.” We learned about the evolution of digital projects and the problems that occurred when new platforms and software required entire projects to be redone. That reinforced my careful approach to digitization. With Shared Shelf, we’re taking our time and making every effort to set standards for future projects.
Now that the selected JFA materials are published on Shared Shelf Commons, we have to ensure that these documents pop up when people search for them. That means creating metadata. Broadly speaking, metadata is “data that provides information about other data.” Each image in the collection needs to be carefully described consistently and accurately using the appropriate metadata scheme. Generally speaking, entering metadata is not at all difficult. Shared Shelf provides a friendly, customizable cataloging environment that requires no knowledge of code or more advanced cataloging procedures. It looks like this:
But the decisions we make now can have lasting consequences. Last year, Danielle the Digital Services Librarian handed me a hefty guide and asked me to pick one. This was no small matter. Think of it like this. How would you describe this image?
Is it a statue or a sculpture? Is it made from stone or granite? Is the main figure a woman or a female? Is she holding a sword, a longsword, or a two-handed claymore? Is the image itself a photograph or a .jpg? And on, and on, and on.
We ultimately decided to go with Dublin Core, a scheme devised in 1995 in Dublin, Ohio for the Online Computer Library Center (OCLC). Dublin Core is relatively simple and allow us to migrate the Shared Shelf collections to Omeka or the Digital Public Library of America. But if we want to jump on the Illinois Hub of the DPLA, we have to modify that Dublin Core a bit and ensure that we conform to their standards as well.
Even before we uploaded the first image to SharedShelf last fall, we decided to publish all of the images as we went. This means that if you really want to geek out, and you happen to be on the Illinois College campus, you can watch the SharedShelf project evolve in real time.