When I started working in the Notre Dame Archives in 1983, we relied on technology developed in the 19th century or earlier: the typewriter, telephone, photography, snail mail. We kept typewritten inventories in file cabinets. We responded to reference requests by mail or by telephone. We had card catalogs with typewritten descriptions of documents. This past June, I met with several of Notre Dame’s archivists to talk about the changes they’d experienced over the years in their respective areas.

How has archival work changed in the last 40 years? To begin the discussion, Head Archivist Patrick Milhoan invited responses to that question.

Senior Archivist Erik Dix recalled that when he started at Notre Dame he relied on an early word processor, but soon had to learn to use Encoded Archival Description, a new XML standard in the archival profession. Appointed as the archivist in charge of audio-visual material, Erik had to make media in many obsolete formats available for current research: “I remember when I started here, we were copying film to VHS tapes to make that available, and now most of the time it’s digital files. With audio it’s similar. We had a double cassette deck so that we could copy cassette to cassette and send people cassettes.”

“We had content within our collections that we would transfer to whatever media was easily consumable at that time,” said Patrick. “It essentially was analog to analog, and you mailed a physical object to somebody. Now we still have the analog material but also the knowledge and wherewithal to convert it into a digital file.”

Erik mentioned the challenges of preserving data from old floppy disks. Archives Specialist Jason Kauffman said: “I don’t have any personal experience using five-and-a-quarter floppy disks, because by the time I came along it was three-and-a-half-inch floppy disks, and then we moved on to the thumb drives. I have to learn skills that are still relevant to doing work in the archives that most people don’t really use today, unless they’re doing something very specialized, like retrocomputing.”

“Scott [Digital Archivist Scott Kirycki] and Jason are utilizing different tools,” Patrick said, “to ensure that integrity and authenticity are still there within the digital field, just as they would be on the analog side.”

“The workflows that you can implement in your repository depend on the infrastructure—currently existing infrastructure in terms of storage or various tools that are available,” Jason said.

“We have a large selection of the old media readers,” Scott said. “Floppy drives of various sizes, Zip drives, the connecting cables that are needed to connect them to a modern computer or additional pieces of technology to bridge the connection from old technology to new. We need specialized software that can read the formats that the older material was stored in.” He described the preservation process for digital files and the importance of standards. “We have storage systems where we’re creating checksums or doing other computational, mathematical things to verify the files to make sure of that authenticity that Patrick alluded to.”

Senior Archivist Elizabeth Hogan is in charge of photographs. She shared her thoughts over email. “I typically am helping researchers with their requests,” she said, “either answering emails or digitizing material. Once the material is digitized, I clean up the scans by straightening and trimming and doing minor color adjustments. The thing that takes the most time in the process is writing the metadata and managing the files in the workflows toward preservation and access. The metadata is usually more descriptive than what’s in the finding aids, so I might have to spend some time doing research on it.”

Access, data, and the limits of digitization

“There’s always going to be something new,” said Patrick. “There will be a new expectation for levels of discovery, new expectations for levels of access to content. It’s absolutely critical for us to maintain a presence within the professional field to ensure that we can always stay in alignment with what’s happening to be able to provide a high level of customer service for the collections that we steward.”

“In addition to all the stuff we’re doing internally to manage this material,” said Scott, “we also have the component of making it accessible and serving research requests or servicing campus stakeholders who need access to the material.” But the archives cannot make everything digitized available to the general public. Scott mentioned the “possibility of things being reused inappropriately—or there are privacy concerns.”

Patrick said, “We describe our content utilizing Encoded Archival Description. We present that information on the web through a program called ArchivesSpace.” The description “is readily available to anybody at any time, but there could be a bit of a user challenge in that you might see something presented online and then expect to be able to click on it and have that content be immediately available. That’s not necessarily true.” Why not? He explained that contracts with donors or university regulations restrict certain records for a period of time. Sometimes people have to come to the archives to view or listen to digitized material.

Scott added, “I think it’s worth mentioning too that, yes, a lot of our material has restrictions, but there is also material that was published—like university publications—that we have made efforts to make available freely on our website.”

“We welcome researchers from all levels,” said Patrick. “It doesn’t necessarily have to be somebody that’s looking to write the next great monograph or somebody that’s looking for an article. Anybody can come to the archives.”

“As for photographs specifically,” Elizabeth said, “I started here early enough to see the impact of social media on the demand for material. Suddenly, everyone wanted a cute photo for Throwback Thursday and the University Archives was the place to go.”

I mentioned that people have sometimes asked me, “Have they digitized everything in the archives?”

“It’s unrealistic to think that any repository could ever digitize the entirety of their holdings,” Patrick said. “I think in images alone we’re in millions. And there are issues related to copyright—there are so many nuances to what we can make freely available at any given time.”

Regarding audio and video, Erik said, “The things that get requested the most have been digitized, and the most fragile things have been digitized.”

I asked what advantages computerized finding aids have.

“ArchivesSpace allows for a lot more data to be recorded about a collection in its corresponding finding aid,” said Elizabeth. “Before, we had a very simple system that had five fields: Collection, box/folder, description, date, and format. Now we essentially have unlimited fields and more opportunities for linked data with subjects and formats . . . The finding aids are online and have been scraped by search engines, so people can find material through web searches.”

Thinking of earlier archivists at their typewriters, Scott pointed out that much of the work archivists do still involves typing at a keyboard. “Before you can manipulate the data in a system, you need the data, and there’s still very much a human element that’s creating that data.”

AI in the archives?

As we approached the end of the hour alloted for our discussion, I asked, “Does anybody want to tackle the question of artificial intelligence in the future of the archives?”

Patrick said, “I think there are significant privacy concerns with this, but I think if you can harness it . . . there would be immense value to the archives profession.”

“AI will be able to help us to describe more data more quickly,” said Elizabeth. “For me, things like facial and object recognition would be a game-changer. I theoretically load thousands of football photographs and AI could tag the coaches, cheerleaders, band, and Irish Guard for me, if I train it to know what those things are and to look for them. If AI could better decipher handwriting, that would be invaluable as well. Having OCR [Optical Character Recognition] on publications has made them so much more accessible to patrons, and we’ve found many important and interesting articles that we would not have found with traditional indexing where only the title and author are noted.”

Erik said, “Digital asset management software now has AI features built in . . . it can be used for very good causes but I’m sure it can also lead to abuse.”

“I think,” said Patrick, “the only way that we would be able to meaningfully use it is to set the standards for adoption and use. Without doing that, and being very transparent in how the field in general does it, you’re not going to see it widely adopted. I think the field itself needs to develop a shared understanding of the implications of it and develop some expectations regarding adoption and use.”

I used TurboScribe.ai to transcribe the audio recording of this discussion.

Wm. Kevin Cawley retired in 2019 from his role as senior archivist and curator of manuscripts at the Archives of the University of Notre Dame, after 36 years of service. He serves as an archivist for Notre Dame’s Jacques Maritain Center and as chair of the digital access subcommittee for the American Theological Library Association’s CRRA program.

Some transcribed remarks in this piece have been edited for length and clarity.

This column appears in the fall 2024 issue of the American Catholic Studies Newsletter.

Originally published by Wm Kevin Cawley at cushwa.nd.edu on October 18, 2024.