In our latest 2015 In-Depth Interview with Library Journal Movers & Shakers from academic libraries, sponsored by SAGE, we spoke with Jason Clark, associate professor and head of library informatics and computing at Montana State University Library, Bozeman, MT. While working as an assistant at Marquette University Library in the mid-1990s, Clark recognized the potential of the emergent World Wide Web to change the ways libraries shared and accessed information. An early adopter, he began writing code prototypes for library systems, and over the past ten years has enlarged his library’s computing team from a solo gig into a busy eleven-person department.
LJ: What exactly do you do as head of library informatics and computing?
Jason Clark: Digital archiving, digital collections, data preservation, and data management are our big focuses right now—but that really doesn’t tell you what I do. These days it’s a lot more about connecting people. I meet with everybody on the team at least once during the week and talk about current projects, how I might help them, where they might need new partners. But I to keep my sanity, because I am a developer at heart, I do a lot of prototyping ideas that are potentially of use to libraries—like semantic classification, text analysis, helping public services build research interest, and connecting people to students.
I’m not the metadata department but I love metadata, so we do a lot with structured data and figuring out how to describe people. We did an application with our staff directory which helped map out topics of expertise and specialties for the people in our library, and make that web accessible and machine readable, which is something I’m really interested in.
In this interview series, sponsored by SAGE, LJ goes in depth with the 2015 Movers & Shakers from academic libraries, delving into just how and why they pulled off the projects that brought them recognition as innovators, change agents, and more. For a deeper dive into what made our 2014 Academic Movers so exceptional, download our 20-page collection of insightful interviews.
Do you also teach?
Yes—lately I’ve been doing a lot of teaching on web-scale cataloging and on interface design. I just published a book, Responsive Web Design in Practice (Rowman & Littlefield). The core of the book is a set of recipes and exercises: How do you build a library web form so people can sign up for certain services or talk to certain people? How do you build a website responsively? How do you build a search interface responsively? You can walk through and learn as you’re working through the book.
You have gotten yourself positioned to do exactly what you want within the library. How have you achieved buy-in, especially when you were first developing web services for libraries?
Early on there was this perfect symmetry: the web was coming up, I was entering library grad school, and I fell into a unit with the division of information technology at Wisconsin and they wanted me to learn web application development. When I came out of that fellowship, I was well equipped to walk into a library environment, work through an idea—not just abstractly, but get it into a prototype mode—and then start to shop that idea around early and get fast reactions.
When I got here I was the digital initiatives librarian—a team of one—and a lot of my work was trying to find partners. I followed a lot of different threads on the open web, not necessarily inside of libraries, and saw things that might be disruptive—what the values were, how they were changing things, sometimes for better, sometimes for worse. And I thought about library processes in the big picture—how they might not necessarily be disrupted, but how they might be improved. When I found that I could prototype around an improvement, the partnerships just fell into place. Even if it’s just a broad-stroke sketch with a web application, people can immediately see the idea and they can get on board.
So I did a lot of that early on. Outside the library, I worked with the school of film and photography on a number of digital sound and film archives, because I saw they had a need for that kind of work, and that ended up being nominated for a Student Webby Award and we got to go to SXSW. It was fun.
I tried to sketch ideas with code, and that seemed more successful than when I just went in and said, “I have this idea, and we should work on X together.” Those conversations were okay, but they didn’t get us to the next level.
What projects do you have in the works?
A recent project I’m starting to think about is the idea of a nanopublication. If you think about the way that scholarly communication articles are written now, there’s a genre to writing. That genre comes out of 19th-century models, and I wondered if there’s room to re-engineer the idea of what an article is, and think through what types of nanopublications you could put out in a new format that would still present the idea, but not present it in 20-page articles—present it in two pages. You’d still want it to hit all the markers of value and quality that the academy has, but I think you can do it in a shorter genre.
I’ve also started thinking about what’s next for search—conversational user interfaces and smart interfaces.
Can you talk a little about your linked data app, GetSEMantic?
When you talk about web-scale cataloging, which I think is an era we’re entering—where we’re trying to describe discrete pieces of web pages, to define people in our organization, to define the services our organization provides—that kind of activity needs what I would call on-boarding or jump start applications like GetSEMantic. You hand GetSEMantic a URL and it runs through an algorithm based on text analysis and helps classify that page, and you get a set of linked data terms and subject and topic terms that you can use to then classify that URL.
I find ways to prototype it and see how best it might work, and try to involve other people in the organization like the metadata group and say, “What do you think of this tool? How might you use it?” That’s my typical workflow: brainstorm an idea, and the brainstorming is actually building something, and then I’ll take that prototype to a potential client or user and say, “Are we on the right track here?”
You cover a lot of bases—access, discoverability, data visualization, linked data—what do you think is the most important computing issue libraries are facing right now, beside web scale cataloging and search?
I think it’s privacy. We’re going to need to balance privacy and customization—I do think we compete with the expectations that Google and search engines have put in place. One of the ways they’ve improved their service is knowing who you are, what you like. Balancing the right to patron privacy with the ability to customize services is going to be in tension, and we’ll want to do well on both accounts—anonymizing data to the point where we can still use it effectively in library applications.
I also think about what I call the tyranny of the algorithm—[where] a filter is being introduced because that system is learning about you. Another interest for libraries should be: How do these algorithms work? How do we make them transparent to our patrons? Because I think that’s a form of information literacy.
You do a lot of forward-facing work with different constituencies within the university. Does your message change, or is it consistent across all of them?
When I go outside [the library] there’s a lot more advocacy. I’ll get into some situations where people are [asking], “Why’s the library doing that? Why would you be doing linked data? Why would you be doing semantic web identity?” Not to come to the table and be brash or anything—it’s just: “Yeah, we do this stuff. There’s a component of us that is interested in the same things you are.” I’m pretty passionate, and if I can show that passion without being over the top, people respond to it. They can see that I care, that I’m engaged, and that goes a long way to building partnerships.
If you had to give three tips to someone in an academic library who’s interested in coding, who has ideas, and who wants to lead, what would you tell them?
- Find your niche—find a strong developer community. Code4Lib is a very welcoming community, they’re interested in people learning, and they’re [interested in] code, libraries, applications, where we go from here.
- Start to publish your code, even if it’s just on your website. Find ways to share it. Typically people are using GitHub. Use that resource to build not only trust in what you’re doing, but also transparency, just putting something out there that people can see or respond to.
- Find a champion or two inside your library. I have really strong colleagues not only within the library informatics and computing department, but in the library in general. Find somebody you can have lunch with, maybe every month, to talk about your goals. There’s usually someone in the organization that has similar interests—create a local interest group.