There are many steps involved in creating and publishing a new digital collection — it’s truly a team effort that requires a lot of hard work and coordination of efforts from people across the libraries, with many different skill sets, working in many different departments, in buildings across Duke’s campus.  People who aren’t familiar with the process often think that digitizing the materials is the most time-consuming part, and that once that’s done, the collection is ready to go.  The truth, though, is that our colleagues in the Digital Production Center, who do the digitizing, are so fast and wildly productive on their scanners and cameras that the rest of us are constantly trying to catch up with them.

One of the most time-consuming parts of the digital collections process, and the part that people often don’t think about, is creating the metadata.   Metadata is data about the materials we’ve digitized, and as part of the metadata process, we have to decide how to arrange the items in the digital collection, how to describe them, what information we need to collect about them, what kind of terminology to use so people can find them, and all sorts of other things.  We have to decide how we want users to be able to find and interact with the digital objects, and what metadata is necessary to make that possible.

To make things even trickier, not only is metadata perhaps the most time-consuming part of the process, but up until this point we’ve had only a small number of staff working on it.  Part of the problem has been that we haven’t had a good metadata creation/management tool, so the workflows and procedures we’ve concocted to get around that have been so unwieldy that it just didn’t make sense to throw tons of staff at them.  But now that our new metadata editor Trident is getting closer and closer to becoming a reality, we can finally think about bringing nearly all our catalogers and archivists into the metadata process, which has been our goal all along.  In early May, we brought two trainers in to teach a two-day metadata course for about 20 of our catalogers, archivists, and other staff to prepare them to do this work.  We’ll soon be putting a subset of that group to work on the huge Broadsides project we’ve been talking about elsewhere on this blog, and then once we really get going, we’ll bring even more of them into this project and others.

Our goal is that digital collections work will become just one of the many things our catalogers and archivists do as a regular part of their jobs.  These folks are already experts at describing, arranging, and providing access to the library’s collections, so now they’ll be applying that expertise to new types of materials.  Even if they only work on digital collections as a small part of their jobs, bringing all these new staff members into the process will allow us to create metadata — and therefore create digital collections — much faster than we ever have before.  And that means more images, more text, more audio, more video … more ideas and discoveries will be possible for users around the world than ever before.  The best is yet to come ….

Some of my favorite objects in Duke’s digital collections are images of children.  Some of them are touching, some of them are funny, and some of them are just plain cute.  But when I see images of children, like this one from the Michael Francis Blake Photographs, I wonder: Who are they? What were their lives like, and what became of them?

There are great images of children in many of our digital collections, including the Michael Francis Blake Photographs, the Sidney D. Gamble Photographs, and Ad*Access.  We’ll soon be adding to this list with a new collection, Images of Mainline Protestant Children and Families in the U.S. This collection, coming out of the Duke Divinity School Library, will consist of photos from American magazines of the mid-20th century, depicting what children and families in the U.S. looked like — or, often, idealized versions of what the creators thought they should look like.

As we build our digital collections, we keep in mind how they work as individual collections as well as how they work together.  By using consistent metadata across collections and developing ways to display and let users interact with objects from many sources, we try to provide seamless access across collections and create opportunities for interdisciplinary research and interesting discoveries.  A search on a term like “children” will bring back images, texts, audio, and video from around the world, various historical periods, and all kinds of social contexts — give it a try and see what you find.