- Spellbound Blog - https://www.spellboundblog.com -

SAA2006 Session 305: Extended Archival Description Part II – ArchivesUM

The second part of SAA2006 [1] session 305 (Extended Archival Description: Context and Specificity for Digital Objects) [2] was a presentation titled “Understanding from Context: Pairing EAD and Digital Repository Description”. Delivered by Ann Handler and Jennifer O’Brian of the University of Maryland [3]‘s ArchivesUM [4] project, this talk explained the general approach being used to tackle the challenges of managing item level description of digitized images at a large, diverse institution.

They described the tension between collection level descriptions and the inheritance of attributes by items. With more images being digitized all the time, storing the images on a network drive made them hard to find or inventory. Different departments at University of Maryland described images at different levels and using different software. While one department had just folder collection level descriptions, another set of 3000 photos were described at the item level but without adherence to any standard.

A flat system of description was not the answer because it provided no way to include hierarchical description. Their answer was to combine archival description via finding aids with well structured item level description.

As the ArchivesUM project defines them, an item can be part of more than one collection and could be part of a collection that represents the source of the item as well as an online exhibition. The team also wanted to create relationships between the existing repository of finding aids and new digital objects. In order to accomplish everything they required (and in contrast with the Archives of American Art’s choice of ColdFusion), ArchivesUM selected FEDORA [5] – an open source digital object repository for their development. They created a transitional database using SQL Server to track and store the metadata of newly scanned digital objects to not loose anything while the custom FEDORA system was being developed. The digital object repository uses a rich descriptive standard based on Dublin Core [6] and Visual Resources Association (VRA) Core [7] .

In order to connect to other kinds of objects from non-archival collections and other sources they simply add them in finding aid. To present the finding aid they needed new style sheets. Their improved layout helps user understand the hierarchy of the collection and how to find the individual record they are looking for, lets them move up and down through tree easily to explore and helps user understand the context of the record they are viewing.

One of the questions that was asked at the end of the session related to the possible drawbacks of linking just one or two digitized documents to a finding aid. The concern was that it might mislead the user into believing that these few documents were the only documents in the collection held by the archives. The response was strong, asserting that the opposite was true – that having the link into the finding aid gave users the context they desperately needed to understand the few digitized records and point them in the right direction for finding the rest of the offline collection. This is especially important in the Google universe. Users may end up viewing a single digitized record from an online collection and, without a link back to the finding aid for the collection, not understand the meaning of the record in question.

The ArchivesUM team is currently setting the stage for migration to the live system. I look forward to exploring the actual user interface after it does.