Shared data dictionaries and sustainable collections | Discussion | Digital Objects in the Classroom
I wanted to repost something I just wrote on the Costume Society of America list-serve, which I thought might interest some of you. My email is in response to Jacqueline WayneGuite, the Emily Reynolds Historic Costume Collection Manager at North Dakota State University. She was asking if anyone had developed a data dictionary for costume for ContentDM.
Don’t just skip over this as being only about costume – the issues considered here apply to any collection of any kind of material, and are particularly important if we want to build large aggregated collections with standardized collections that can be easily searched by our students. The issue of sustainability is also important if we want our classes to be able to rely on continued access to a resource.
Here’s my reply:
I know others have worked on this (but I’m not going to volunteer anyone else!), but with our collection at Vassar we haven’t created much of a controlled vocabulary yet. Part of the reason for that is that I’ve been waiting for a time that several of us could work together collaboratively to develop one that could be agreed upon and shared consistently by the costume history community. This could be developed as a part of the Getty Art and Architecture thesaurus (used by many visual resources libraries) or as an alternative to it in our field. I always have my eyes open for grant funding opportunities for working on such a project – if anyone else would be interested, let me know and I’ll look more actively for a collaborative grant application. We need one shared resource for best practice in digital costume collections.
If anyone does have a data dictionary they’d be willing to share, that could be a great starting point for others to evaluate and build something standardized. So far I find that everyone builds an individual dictionary for their collection, and since specific collections (and our own use of vocabulary) are so diverse, they aren’t very standardized. My dream is of an aggregated resource where we can search many of our collections at once, so you can imagine standardization is important. Imagine being able to search costume collections all over the country the way you can search in library collections!
As for ContentDM, I would recommend that you move forward with careful consideration. The library at Vassar encouraged us to use Content DM 8 years ago when we were starting our digitization project, but luckily we were slow in moving forward – 8 years later ContentDM is barely supported on our campus and another product, Luna Insight, has overtaken it. Sustainability of digital collections is a big concern. They key is to keep your own data somewhere separate. Do you have your own private database already for the collection? Or even something as simple as an Excel spreadsheet? What we have done at Vassar, as many librarians have recommended to me, is to keep our own local database (in our case, Filemaker), which then can be imported into another product such as ContentDM or Luna Insight or Omeka. Hopefully by now it’s possible to export out of ContentDM and back up the data, but I’d rather have a backup of my data in the beginning than in the end. We’ve worked so hard on collecting our data, I’d hate to leave it in the hands of a librarian or IT person who may move on to another project and leave me in the lurch (it happens!).
I look forward to hearing responses from others on this issue. I’ll post this on a few different lists myself, to try to keep the conversation going.