[Edit: added link to online workshop materials]
Usually I spend my Fridays working at home trying to stop my cat from encrypting my e-mails, but Friday August 25th I spent at the DLF Open Archives Initiative (OAI) Implementer’s Workshop (I’ll link to handouts once they come online) at Stanford. During the general introductions, I declared my purpose at the workshop as “to poach as much information as I can for the museums I’m working with” (and in particular the Museum Collections Sharing Working Group), and subsequently I found out that a lot of the instructors and participants had interesting stories to share about OAI and the museum world.
Tom Habing from UIUC commented that he was working with the UC Berkeley Art Museum on OAI harvesting for their Museum Toolkit, a FileMakerPro database recently renamed from DAMD to DAVIS (I wonder whether it had anything to do with propriety?). He also thought that this tool would include a data transformation to CDWA Lite XML (I’d still like to confirm that with my former colleagues at BAM).
Martin Halbert told me about a tool called “Metadata Migrator,” a tool developed during the project “Music of Social Change.” This project involved various museum / historical societies / archives who couldn’t run their own OAI servers, but using the migrator tool they could map their local data to DC and automatically create OAI-compliant XML records (I assume in the form of a static OAI repository; see below). It also yielded a report [pdf link] (which I have printed, but not yet read), which details the museum’s OAI experience.
I also learned about so-called “static repositories,” which consist of one long OAI compliant XML file – once these files get registered at a static repository gateway, they, too, can be harvested and shared. For small institutions not capable of running a server themselves, this would be a way lower the threshold for participation.
I also gathered that a reasonably big leap in the further evolution of OAI must be a clearer way to demarcate what sits at the end of an identifier URL. Harvesters can’t really make much hay of the URLs if they don’t know whether these lead to a webpage about the collection the resource is a part of, or an image of the described resource in context, or a full-resolution image, or a thumbnail. The “Asset Actions” developed during DLF’s Aquifer project are a step towards resolving this issue. While OAI traditionally has been purely about sharing the descriptions, clearly interest in sharing the digital content along with it is on the rise.
Kudos to the workshop instructors Tom Habing, Kat Hagedorn, Martin Halbert, Liz Milewicz and Jenn Riley, and to DLF for pulling it all together! The format of the workshop (some introductory lectures & small themed break-out groups for the rest of the day) worked really well for me, and I thoroughly enjoyed all the interactions between the instructors and the participants. In oh so many ways, a big improvement over fighting for the keyboard with my cat…