Librarian Dean Seeman makes metadata matter

 Librarian Dean Seeman

In your role as Head, Metadata at UVic Libraries, what does your work involve?

I oversee the description of resources so students, faculty, staff, and the general public can reliably find and access library resources. Traditionally this has been called cataloguing but the more inclusive term “metadata” has been adopted to indicate that this can include description of resources in other systems and outside of the catalogue.

A lot of this work is standards-based so it requires familiarity with a wide range of standards (stacked on top of each other they would probably be as high as the library itself!) and then we have to make decisions locally of how to apply those standards.

My work is to know the standards and related issues well enough to set strategy and make overall decisions, and then manage and implement these decisions among the great team that does this work. We have specialists in digital collection metadata, special collections and rare book cataloguing, music, law, serials, and non-English languages.

What kind of comments or questions do you receive from colleagues regarding the process of cataloguing library material?

Usually it’s questions about why something appears in the catalogue a certain way or why a piece of information is missing – questions that require digging into what the standards dictate, how we are interpreting the standard, and whether a piece of data is missing or should be present and visible. We also push descriptions from our various systems (including the catalogue) into Summon, our over-arching library and archival discovery tool. This generates a lot of questions about mechanics of moving data between our systems. Since we also deal with technology quite heavily, I also get questions on the use of technology for metadata and cataloguing work.

Additionally I get asked for advice on how to format metadata for digital projects. For example, if someone is digitizing a set of photographs and wants to make them available online, I would help with how to describe them consistently. I would help them work through what information to include, what standards are available, and how should the metadata be formatted so that users can search, filter and explore the collection.

Your team was involved in the launching of a new online discovery tool for The Transgender Archives, one of the largest collections of trans archival material. Share with us what went behind creating this tool.

We have had a role in a couple of projects related to the Transgender Archives. One was to help set up a framework for community members to describe transgender periodicals in our collections: What articles are in it? Who wrote them? Who took the photographs? Who edited the issue? What events does it list and where do those events take place? Volunteers and other contributorsfilled in the data overseen by Michael Radmacher and Aaron Devor. Much of this data was also made available for visualization which required further standardization: work done by Michael and Data Curation Librarian Shahira Khair. Our work in this project (and others) is to help set up a framework for collecting good, relevant data, answering questions along the way and assisting in any other way we can.

The Metadata Unit has also been involved in describing the Transgender Archives material that has been digitized: namely Transvestia. Like other resources we describe, we provided the metadata for each issue in this collection.

What projects are you and your team currently working on?

Besides our day-to-day description of material, we are involved in a number of exciting projects. Really every new digitization project and its metadata is a different challenge and is endlessly interesting.  We are also currently migrating to a new digital library platform which keeps us busy. This new platform, Vault, is intriguing particularly in its use of linked open data to describe collections. Tim Berners-Lee proposed the linked open data concept as the next evolution in the Internet and allows us to think about our descriptive data not just in how it operates within our local databases, but how it can interoperate with other data and metadata from around the world.

To that end, we are also working with the amazing team in Special Collections and University Archives to describe our archival holdings in greater detail. Related to this is work to take some of our data that only exists in textual form, and add it as discrete data to Wikidata – a sister project of Wikipedia that is the world’s foremost storehouse of structured data. This involves transitioning our thinking from the description of objects to the creation of knowledge.

Your work involves working closely with other departments at UVic Libraries and colleagues around the world. What best practices are we doing that other academic libraries may be interested in?

I think metadata is often viewed in silos across an institution. I usually argue for a common approach to metadata across all systems so that we can provide consistent discovery. This isn’t always possible but we at least strive for the data to be complementary. Metadata is often an afterthought to a project or in the course of regular work, but it makes everything possible. An organization or project can fail due to poor metadata. There is a saying that “metadata ages like fine wine, systems age like fish.” In other words, a website or catalogue will have to be replaced as technology and user preferences change. The underlying metadata, if done well, has a much longer shelf life that usually outlasts its system.

Knowing the importance of metadata, I also think it’s incumbent on us to change our metadata and cataloguing practice to better correspond with the way the world describes and consumes information. Since the advent of the Internet, the library is no longer the lone gatekeeper to trusted information – so the question becomes how can our knowledge and expertise be leveraged in the Internet environment. As far as metadata goes, this pathway I believe is provided through linked open data.

It’s also very easy to get lost in minutiae when dealing with metadata and standards. But when I walk into the library and through the stacks, I don’t see books but instead I see people, places, events, material objects, and ideas in conversation with each other. I see the role of metadata to reveal those conversations so that people and machines can also enter into that realm, discover each other, and enter that conversation.

Metadata is usually not noticeable but it is the primary way of communicating on behalf of a resource. Within that is a depth of work that attempts to understand authorial intention, and line that up with the language that a standard or the public may use to represent that same idea. The work is really endlessly fascinating. It requires coming to terms with an idea (in physics, literature, mathematics, chemistry, sociology, musicology, etc.) and trying to faithfully represent the author’s intention, the idea itself, and how you think people will want to find it.

If you could “take over” the UVic Instagram account, which item from the collection would you want to show their followers and why?

I had the good fortune to spend a couple of years in northern Japan (Aomori) so the photographs from the Herbert Geddes collection are particularly interesting to me. These images have been digitized from glass slides created of scenes in Japan in the early 20th century. I would choose this image.

It takes me back to the Japanese countryside where I lived nearly a century later, and where you would still see people working in rice fields throughout the year.

What technology from science fiction do you wish existed today?

Definitely a sarcastic robot sidekick.

Interview conducted and edited by Zehra Abrar.