Interview with Emily Baldoni, May 24, 2021 (July 13, 2021)
Sam Smith
July 13, 2021
Emily Baldoni works as the Metadata Librarian for Digital and Special Collections at Lauinger Library at Georgetown University. Samantha Smith, who works as a research assistant for the On These Grounds project, sat down with Baldoni virtually near the end of May to discuss her work on the On These Grounds project and how it relates to her role and responsibilities as a Metadata Librarian. This blog post is a prose-style transliteration of their interview style conversation, written from the perspective of Smith listening to Baldoni’s responses. Words in quotation marks are Baldoni’s, taken from Smith’s notes. Phrases in italics denote specific questions posed during the discussion.
Our conversation started with the question, what challenges and opportunities arise in thinking about data models through the lens of events? Baldoni’s metadata background offers a “unique approach” to the work of the On These Grounds project, as she is not an archivist or historian. “Linked data is really at the cutting edge of many of the new metadata initiatives in the library community” Baldoni said. “But many of the linked data standards being developed in the library world are essentially bibliographic or document-driven data models,” she continued, that focus on using linked data properties to describe books, documents, or other items in a library's collection. For the On These Grounds data model, in contrast, “the unit of analysis is not a document, broadly construed -- it is much granular. It's the individual events found within the documents. It's really a different paradigm.” The description and implementation are also incredibly granular.
Baldoni mentioned that the On These Grounds model is “different from the descriptive approach found in archival contexts as well” in part because of the professional impact of MPLP (more product, less process). While MPLP has helped open up broad collections to researchers and reduced processing backlogs in archives, Baldoni noted that it is “difficult to make individual events at the level of detail we are interested in discoverable, accessible, with an MPLP approach alone.” To do that, specifically with the work for On These Grounds, you need to “dig deep down in the document to get at the lives of the people” which “requires much more granularity. In a sense it's at the opposite end of the spectrum from the MPLP approach.”
From Baldoni’s perspective as a metadata librarian, On These Grounds as a descriptive model “brings together description of archival collections with metadata technologies that been developing on the library side in a really interesting way.”
Working on On These Grounds has given Baldoni a chance to “work closely with archival collections that wouldn't normally be part of my daily work.” She noted that she is “grateful to have Mary Beth [Corrigan] on the team” because of her “deep understanding of collections and history of enslavement and its archival legacy.” Baldoni and Corrigan “bring different backgrounds to it” and Baldoni “can’t imagine working on it without her.”
Considering the different approaches, backgrounds, and perspectives that Baldoni and Corrigan bring to the On These Grounds Project, lead to a discussion about Baldoni’s role and responsibilities as a Metadata Librarian. She does “cataloging and metadata for rare books, as well as digital materials in Georgetown's institutional repository.” This can include archival collections that have been digitized, but it entails a “different way of interacting with archives” because “by the time the collection is digitized, the finding aid has already been created. I use the description in that finding aid as the baseline, deriving as much metadata from the description already created as possible” and then enhancing it for re-use to describe individual archival folders that have been digitized.
For her work on the On These Grounds project “the descriptive approach is very different” because the data being recorded is so much more granular than the collective description used in a finding aid. However, Baldoni stresses that the On These Grounds model is in no way a “replacement for a finding aid.” It is complimentary. When asked about the challenges and opportunities of the descriptive approach, Baldoni said, “the challenge for me personally is that I don’t come to this with pre-existing expertise about these collections. So the context provided by the finding aid is all the more important.” Through her work with the On These Ground project she is “learning about collections and how they are structured.” “Context is essential,” she continued, “you need additional information about the collection and the circumstances in which it was created in order to infer things about the events being depicted within individual documents.” And, she added, it is “not all wrapped up in a bow. [There] [is] no bow. It can take interpretation, and that can be hard.”
Another challenge is the content. The experiences of enslavement are traumatic. In the description of events, Baldoni said deals with the “dehumanization of people because they are referred to as property, treated as property.” It is not “easy material to work with” Baldoni said, but the work itself although difficult is important.
This lead to a discussion about secondary-trauma at work, and how she deals with it. Baldoni said she sometimes has to “take a break from it” because “you don’t want to desensitize yourself.” “We need to do that work to make those aspects visible,” and we are “doing this to make that history and those experiences available.” This is where it is hard she said, “because we want to record information from these collections in a way that is structured and can be queried, but at the same time, you don't want to perpetuate that dehuminization by reducing people and their experiences to mere data, to numbers. It cannot become just a way to test a data model.” Thus, it is important to “keep in mind why we are doing this” and that it is “not just data.” It is “data related to the experiences of real people.”
The conversation then turned to the pandemic, and specifically how it has impacted the project. “In many ways,” Baldoni said, “for me it freed me to do more on this project.” This was in part because of the timing, but also because so much of her work with physical materials in the library deals with the backlog of rare books. Thus, she “was able to use that time in other ways [because] the project came at a time when we weren't in the library and couldn’t do physically based tasks.”
“We all meet over Zoom,” she said of the On These Grounds project team, and we “meet every two weeks.” We have not met in-person, and this “puts different strictures on how you collaborate.” “Working with people over Zoom has become very normalized because of the pandemic,” she continued, and “things that we might have viewed as a barrier to collaborating before, like being physically located in different states, at different institutions,” now are not seen that way.
As she reflected on the testing process over this past academic year, she said “it’s been so interesting to be involved with different phases of the project.” In creating the data model, to describe events related to history of enslavement at colleges and universities, the metadata categories have been formulated both “top down” and “bottom up.” Baldoni keenly observed that we “look at things we need to describe and then create terms based on the properties of those things, but at the same time it is impossible to start describing something without at least some initial categories in mind.” Within this project “there has been an effort to do both,” meaning create metadata top down and bottom up. “You start with some initial categories, an initial framework, but as you start working with the documents, you find that there are details that escape the terms of the model, and as a result you need to revise your categories. The feedback can make it messy,” she continued, “but messy in a good way. The next round of testing will be messy too, because as we continue to work with a widening range of documents, we will find more things that don't fit neatly, and we can in turn use that feedback to imrpove the model.”
“There is a temptation in my line of work,” Baldoni said, meaning metadata, “to be overly abstract.” This is challenging because you can “create a structure that is perfectly thought out and immaculate, but applying it on a practical level to documents you need to describe is totally different.” Put another way, if it is “a perfect model but so cumbersome that no one wants to use it, it's not such a good model after all.”
Another challenge is “describing everything that can be described versus describing what needs to be described, and what is that, and what is the differentiation?” The next round of testing involves more data, and so she is curious about what happens when the data created with the model is aggregated. She is also looking forward to, “seeing what kinds of questions you can ask of that data.” Before you have a critical mass of data, “it is difficult to know how granular you need to be.”
The data model is a work in progress. “Terms are developing as we apply them,” Baldoni explained, and “we actually created formal definitions after we had already been applying the properties in testing for some time.” This is the “nature of this work, trying to do it organically, responsibly, and deal with a bit of flux.”
For Baldoni, working on the On These Grounds project is a rewarding experience. She is getting to know part of the collections at Georgetown, and she is a “metadata nerd.” “Constructing a controlled vocabulary, taxonomies, I love that kind of thing,” she said. Library work can sometimes be “a little siloed, especially during lockdown, when we were all working from home, so it was really nice to work so closely with everyone on this project.” Everyone on the team comes from “different backgrounds and different institutions” which makes it “wonderfully interdisciplinary.” It is “really nice to work on something in an un-siloed way,” she said and it is “great to work with such a collaborative team, especially for this type of work.”
The project is coming at good time, as there is a lot of interest in the histories and experiences of enslaved people, as well as the history of institutions in relation to slavery. Baldoni concluded her comments by saying, “I’m hopeful that we are well poised to make something that people can use to work towards these goals.”
To learn more about Emily Baldoni and her work as a metadata librarian, please check-out her interview for the ‘Staff Spotlight’ section of the Georgetown University Library’s newsletter, published on February 8, 2021.