Author, Institutional Repositories, Charleston Briefings Series

Interview conducted by Michael Upshall 

Against the Grain V38#4

This is a lightly edited transcript of a conversation between Michael Upshall, Editor, Charleston Briefings and Guest Host for ATG the Podcast, and Josh Cromwell, who is the author of the recently published book Institutional Repositories, the latest installment in the Charleston Briefings Series.  

Josh C. Cromwell is scholarly communications manager at the University of Southern Mississippi, involved with Aquila, the university’s institutional repository, since its launch in January 2012, and its administrator since July 2013.  He developed and co-teaches an institutional repositories course for ALA Core and is lead organizer of the annual Southern Miss Institutional Repository Conference (SMIRC).

The full interview can be viewed at:  https://youtu.be/v_usIDzZBlw

Michael Upshall:  Hello and welcome to ATG the Podcast.  I’m pleased to be talking today to Josh Cromwell, author of the new book, “Institutional Repositories” in the “Charleston Briefing Series.”  Josh, you’re based at University of Southern Mississippi?

Josh Cromwell:  Yeah, I’ve worked with repositories since I’ve been at the library, but the initial job I applied for was actually a reference position.  This was right at the time that the university was getting ready to launch the repository, so my position pretty quickly transitioned once I got here.

MU:  So your involvement with repositories is quite accidental?

JC:  Yeah, it really was.  In fact, when going through that interview, I did not even know what a repository was. 

MU:  Goodness me!  And the repository was just starting, so presumably nobody had much of an idea about what was involved.

JC:  When I came on board, they had just hired an associate dean who had worked with a repository at a previous institution.  The idea behind her coming here was to spearhead that process of launching a repository.  She had some background;  the rest of us really did not.  I missed the initial planning phase, selecting a platform, all of that business, because I started in November 2011, and we launched in January 2012.  So I did get in on the ground floor of the implementation, the rollout, promoting it to campus, that sort of thing.

MU:  Why join the library in the first place?  What was the appeal?

JC:  I’ve always been passionate about education in general, and libraries merged two of my interests:  it was a chance to do both education and research because my prior background was in history, so lots of time in libraries.  This opportunity presented a way to merge both those worlds. 

MU:  You’ve now been involved with repositories up to the present day — a good 14, 15 years?

JC:  Yeah, going on that this November it’s fifteen years.

MU:  Your role at the beginning was uploading content?

JC:  At the beginning mostly that, uploading, doing a little bit of promotion, working with the faculty — a lot of the more nuts and bolts day-to-day operations at that point.

MU:  Over time, you’ve become involved in training;  this is not your first book about repositories, and you’ve also set up a conference [the Southern Mississippi Institutional Repository Conference, known as SMIRC].  How did you move to disseminating and exploring the world of repositories?

JC:  I guess in July 2013, I became full administrator of the repository.  That associate dean I talked about moved on to another position.  Then, in 2014, ACRL hosted their scholarly communications roadshow at University of Mississippi in Oxford.  During that workshop, which of course was scholcomm writ large, so a lot more than just repositories, but there was one chunk in the schedule they’d left open and then said, here’s a list of possible topics.  Which of these would you like to know more about?  The one that won the overwhelming audience vote was repositories.  Looking around that room and hearing some of the different stories, I realized, we’re pretty much the only institution here that has an IR (institutional repository) at this point.  Why aren’t we leveraging what we know to provide some training for these other institutions that are looking to get their own IRs going?  That was the idea.  Our first IR day (we called it back then) was in 2015, geared toward other institutions in the state to say, hey, a lot of you expressed that you’re interested in doing this.  We’ve been doing it for a little while now.  We’re still fairly new, but we’ve at least got some experience.  What if we brought some folks in for us to learn from, but also for you to learn from, some of our experience, and help you as you progress down that road.  It has just organically grown from there.

MU:  How many people did you get in the first year?

JC:  Maybe ten to fifteen — it was pretty much just our own state. 

MU:  From the beginning, you said the idea wasn’t just to tell others what you did but to learn from what others were doing?

JC:  Yeah, absolutely.  At that point we were still pretty new to the game as well.  The first conference, the keynote speaker we brought in was Marilyn Billings from UMass Amherst, one of the leading folks in IRs in the U.S. at that time.  She had lots of experience, lots of knowledge — we learned a ton from her, just as the attendees did.  We weren’t saying, hey, we figured this all out, come learn from us.  It’s hey, we’re still learning too.  As we learn, come benefit from that as well.

MU:  The principle of IRs was established around the time that Open Access was founded, around 2002.  But when did repositories really get going?

JC:  I would say for at least our area, probably about a decade later is where you really started seeing repositories start popping up pretty frequently.  Most of the IRs, at least that I know of, were in the 2010s, I would say.  We were founded in 2012.  I know of several of other institutions around that were in the 2015-16 range.  But there are still to this day new repositories popping up.  One of the side projects I have is that I teach an IR course for ALA Core, and routinely we’ll have students there who will be taking the course because they’ll say, “Our institution’s getting ready to launch a repository for the first time, and we’re just trying to see what we can learn about them.”

MU:  But most institutions will have a repository today, don’t you agree?

JC:  Right, yes, at this point most institutions have them.  Most of those exceptions I’ve run across are smaller institutions that maybe don’t have the resources available and so a little later to the game than some of the bigger schools.

MU:  Today, repositories are ubiquitous, not just in the States, but around the world.  What do you think has been the reason for such a dramatic shift in library strategy?

JC:  There’s a couple of different levels I would point to.  For one thing, the reason you’ve seen the growth take off the way it has is that it took some of the proof of concept from those early institutions to develop them, get them going, and for people to understand that this is feasible.  In terms of the philosophy underpinning it, it’s that same issue we’ve all dealt with for years and years, which is that getting access to research, especially when it’s behind paywalls, has become financially unsustainable.  If there are other ways we can get access, or provide access, to make that content that we as an institution are generating, to make that scholarship available, and make sure people can actually read it, use it, incorporate their own research, then that expands the institution’s profile.  It’s a way for people to see what researchers are doing to benefit from that research in a way they can’t if all that knowledge is chucked behind a paywall. 

MU:  Do you think it’s true to say repositories have fulfilled their intended purpose?

JC:  Yes and no, I would say.  If you’re looking back at the initial promise of repositories, where the idea was that they’re going to become these complete collections of all of the scholarly output of institution and have that ready and freely available, certainly they’ve not fully achieved that goal because of copyright  and all the different considerations there.  But in terms of really showcasing what the institution can do and expanding beyond that initial vision to now, the fact that repositories are including things like theses and dissertations or other research outputs, gray literature, for example, that wouldn’t fit in some of the traditional publishing profiles, but still has a lot of scholarly and academic import, I would say in that regard, they’ve very much achieved the goal because they’ve become a home for valuable research, valuable scholarship that wouldn’t have a home elsewhere, that wouldn’t fit the profile of some of these other outlets.  So yes and no.  No if you’re going back to just that initial vision, I guess, but yes if you realize that over time we’ve realized that the vision could be a lot broader and more expansive, and  repositories have very much succeeded in meeting those needs.

MU:  That seems to be one of the differences between national styles because repositories developed in the States, as you say, with lots of gray literature, stuff that isn’t captured anywhere else, whereas in other countries, you sometimes get more of an insistence on green OA articles.  Do you think there’s a lesson there for different countries?

JC:  I definitely think it’s a both-and situation, not either-or.  I don’t think that it has to be one approach or the other.  I think that there are ways in the States that we could do a better job with green OA, and I think that perhaps there are ways that internationally the repository could be used more expansively than just maybe some of the green OA focus.  If that makes sense.

MU:  Absolutely.  Well, let’s talk about your new book.  You don’t mince your words describing repositories and what works and what doesn’t work.  In the first place, why did you write the book?

JC:  I’d just edited a book for ACRL a couple of years ago, called Rethinking Institutional Repositories [2024].  That was a collection of vignettes from different repository managers about some of the different projects and use cases they had done, going back to what I mentioned before.  That notion of finding more expansive ways to use the repository to showcase and make available collections that maybe we hadn’t really thought about before.  My book was a way to approach IRs from two standpoints, both the fundamentals, the basics of IR, the different considerations that go into building and growing the IR, leveraging some of my experience of doing this for a while.  But then also trying to look a little bit at that global context and thinking about, okay, 20 years in from where IR started, where are we now, and where maybe could we be headed?  How can we move towards a future that makes the IR even better than it is to this point.  That’s kind of the idea, I guess, to look at both the local picture, the individual IR picture, what goes into building one, as a resource for folks who maybe are new to the space, like most of us that I’ve talked to, including myself, like we talked about before.  I didn’t come into this role that I’m in now knowing anything about IRs.  I had to learn it as I went along.  

Hosting SMIRC for several years.  I’ve talked to many people who have come into the repository space almost accidentally.  It’s, hey, the institution decides we need to have one, and they didn’t really know who to do it.  So it fell to me to do this.  And I had to kind of learn as I went along.  So the goal is to for this conference to be a resource for people in that similar position, to have a good foundation:  You’ve been tapped for this role now.  You may not know a lot about it.  What goes into an IR?  What are the essentials that you need to know?  And from that point, for the more experienced IR manager, you get into considerations like preservation or accessibility or looking at how we network these IRs globally.  How do we make them play nicely together?  How do we build these open systems that work better on a larger scale?  That’s the second half of the book.

MU:  From your experience both as a trainer running courses and as a published author, what has surprised you most from the feedback you get?

JC:  Probably the thing I was just saying, the number of people who have ended up in this space almost by accident, who didn’t really start out down this road but have had to learn this themselves.  That’s been the driving thing that’s led me toward this:  I’m not the only one who has this story.  A lot of us have very similar variations of the same story, where we didn’t go into this with a lot of advanced training or knowledge.  It’s just, here you are, figure it out, go from here.  Hearing how many people had similar stories, but then, once they got into that work, they have seen the value of an IR and have grown to enjoy and love that work and buy into the potential that exists with IRs, has been very encouraging to see.

MU:  The SMIRC Conference has been running for ten years.  Is it free to attend? 

JC:  It is, yes.

MU:  You must have put a lot of time and effort into IRs.  Your book is published Open Access, which for you as an author and trainer means a lot of time and energy.  You are paid for running a repository, but all the training and talking to people and writing the book is something you do off your own bat, as it were — that’s a lot of effort.

JC:  Sure, yeah.  A lot of it stems from going back to what I mentioned before, my background and enjoyment of education.  I just enjoy teaching, I enjoy that side of the process of academia in general.  But it’s also having heard those stories, having heard how many people had my experience, and being able to hopefully alleviate some of that, to give them some skills, to give them some of my experiences or things from other people as well.  Going back to the book I edited, there’s a lot of voices in that besides me.  I don’t want to pretend I’m the one who’s figured all of this out;  I’m just one perspective among many, but to take all those voices from peers and colleagues and be able to share that with others, and to hopefully make that transition into the world of repositories a little easier than it was for some of us who had to just figure it out as we went, is one of the big motivators.  I’ve just wanted to help the next generation of IR manager so they’re not having to learn the hard way, the way many of us did.

MU:  In your book you say institutional repositories are central but underutilized.  What do you mean by that?

JC:  I think that, at least in my experience, and in the experience of others we’ve talked to, a lot of faculty don’t fully understand what an IR is or understand the value it can bring to them.  A lot of that just comes down to that in most institutions you’ve got very small teams working on IRs.  You’ve got the nuts and bolts, day-to-day things that have to get done, and there’s trying to get the word out to everybody;  it can become difficult to accomplish all of that.  In a lot of cases, especially when you’re dealing with small teams that have a lot of work to do, the messaging doesn’t really get out to faculty, or to the campus community as a whole, about what the IR is, and the things that it can do to enhance and showcase their research.

MU:  It certainly seems repositories are a well-kept secret, in that not many people know about them outside the ones who actually work on them.  How many people do you have in your team at Southern Mississippi?

JC:  Currently there are two of us:  myself, and my colleague Jenny, who is the institutional repository coordinator.

MU:  What do you think is the future for the repository?  In your book, you used the term “greenaissance.”  What do you mean by that? 

JC:  I see potentially the chance for a real resurgence or perhaps even, you know, greater than it was before in the use of repositories in general, but including institutional repositories, with some of the changing landscape of scholarly publishing right now.  One of the things a lot of institutions are discovering is that the initial promise of gold OA has not fully lived up to its potential because in a lot of cases all that’s happened really is price shifting.  It’s still that same unsustainable cost you saw in the big deal models, where universities are paying large sums to get access to all these subscription journal packages.  Now it’s that same process, except now, instead of paying for subscriptions, they’re paying for APCs.  That’s becoming less sustainable.  You’re seeing more and more funding agency grants saying we’re not going to pay for those APCs.  You’re seeing that with federal policy in the U.S.  You’re seeing that with things like Plan S.  And so, all of a sudden, if researchers don’t have the finances to pay some of those large fees, they’ve got to find other ways to comply with mandates to make their scholarship open.  In which case, I see the potential for repositories to stand here and raise their hands and say, “Hey, we’re over here.  We’ve been doing this for a while.  We can do this now as well.  There’s no fee associated with us, so come talk to us, and let us help you.”  I see some real potential, especially as those fees become more and more onerous.  And as funders are more and more inclined to say, “Hey, we’re not going to foot the bill for that,” then I think there’s some real opportunity for repositories to leverage that, to help researchers comply with those mandates in a way that’s far more financially sustainable than the APC model.

MU:  And what about AI?  Do you think the advances in AI over the last few years, particularly with genAI, make a difference to the way a repository should run?

JC:  For me, I think the biggest potential for AI with repositories is for some of those day-to-day tasks I was talking about before — especially some of the more repetitive tasks.  I could see potentially AI being helpful for things like metadata, for example.  I could see it even maybe for site design, or building out IR platforms, like the coding side.  To me, that could have some real potential because that gets back into what I was talking about before, the issue of small teams with a lot of workload.  If we’re able to offload some of those more repetitive tasks that don’t require high complexity to AI, that frees up more of the staff for things like the outreach that I talked about before or some of the top-level planning, things to help with networking IRs or making sure they’re interoperable, those sorts of things I’ve mentioned before.  That’s where I would see the real potential, to offload some of those tasks.  Accessibility is  another one:  there’s a potential for AIs to help with things like transcribing audio for lectures or recordings in the IR or even potentially doing alt text for images, or things of that nature that require a lot of manpower at this point to do.  If you’re able to offload some of that and then dedicate your manpower to more of the outreach, more of the promotion, more of the really maximizing the IR’s potential, that could potentially be a very helpful thing.

MU:  One of the things I noticed from talking to repository managers is, as you say, the amazing amount of manual labor involved in getting individual articles into the system and tagging them, adding the metadata.  Don’t you think one of the challenges is that trying to find a technical solution requires skills that perhaps repository managers themselves are unlikely to have?

JC:  Sure, absolutely.  I think that’s one of the biggest barriers you come across.  Or if someone does have those skills, then more often than not, they can find much higher paying jobs to use those skills outside of academia, right?  Finding somebody that has both the skills and the passion for the work and is willing to accept that, “Hey, you’re not gonna make a fortune doing this work like you could in some other industries,” becomes hard to find.  It’s almost like looking for a unicorn sometimes to find that merger of skill sets.

MU:  At the recent Open Repositories Conference, many of the speakers talked about the issue of their repository collection being scraped by AI bots.  Isn’t this a fundamental issue with repositories:  they were set up to make content available, and yet here’s content being accessed and they didn’t always like it.

JC:  Sure.  Yeah, it is.  But I guess my pushback at the same time would be to say that a lot of that content, especially if you’re dealing with published articles, that most of the large publishers at this point have signed agreements with these different AI companies to allow their content to be scraped.  So it’s really happening one way or the other, whether through the IR or not.  I do think there are some considerations to take with regard to AI and the IR for a couple of different things.  Number one, I’ve heard instances of institutions who have basically had their platform being taken down at one point, just because the traffic from scrapers has overwhelmed the site.  So obviously that’s a concern.  There needs to be some ways to kind of manage that.  The other thing is to understand where the traffic is coming from, because so much of the reporting around IRs looks at things like download counts, visits, or statistics, that kind of thing.  Obviously, if those numbers are being heavily inflated by AI, that’s something we want to be able to identify so we can see what our usage is like in terms of actual human beings versus bots.  But yeah, there’s some concerns there and some things we need to consider and think about how to address.  But at the same time, those concerns are far more fundamental in the system beyond the IR itself.  If your concern is, hey, these AI platforms are accessing my work, chances are if you published it, they already are. 

MU:  Doesn’t the issue of genAI scraping content raise the fundamental purpose of the IR, which is to make content available?  One of the things you raise in your book is that individual repositories themselves are not going to get huge amounts of traffic, but it’s vital they make content available.  So, in a way, the creation of LLMs is exactly what institutional repositories were designed for.

JC:  Yeah, I mean in a real sense, yes, because it’s all built on that idea of making content available, right?  And there’s a sense, too, that we come across quite a bit with genAI, the risk of lots of inaccurate information.  Isn’t it better for those corpora, if they’re going to be used in that way?  We can have bigger conversations about kind of the uses of genAI and that sort of thing.  But if they’re going to be used by people to find information, isn’t it better that they’re being trained on solid scholarly research, as opposed to the wild west of whatever’s out there on the internet that could get into all sorts of unfounded theories or ideas or what have you?  Isn’t it better that they’re at least getting access to some sound, peer-reviewed tested scholarship to hopefully provide more accurate information to users?

MU:  Yes, I couldn’t agree more.  I remember going to a conference where an academic said in the space of a few minutes she wanted access to all the world’s scholarly content, but she didn’t want her own articles to be available.  It seemed a bit perverse.

JC:  I do hear a decent amount of that sort of “open for thee but not for me” idea:  I want access to all your stuff, but no, mine needs to be behind this wall because of whatever, you know.

MU:  What next?  Now you’ve completed the book, which is a major achievement, you still have the conference to look after, but what are your next plans?

JC:  One of the things that I’m looking at right now, and it stems back again to that education piece, is the state of IR and scholarly communication education as a whole.  One of the things I’m trying to do currently is explore what sorts of courses are out there for library science students.  Are they getting that background in their educational programs now?  Or is it more things like the course I’m doing for ALA, or events like SMIRC or Open Repositories, where they’re going to be able to find that knowledge?  How has the educational side of the profession met or not met that need for training in IRs or in scholarly communications for the librarians of the future?  And if not, what are some ways to address that?

MU:  How much training did you get on repositories in your own library training?

JC:  Not a great deal.  In fact, quite frankly, most of the time, whenever there are classes here covering IRs, they’ll bring me in as a guest speaker to cover that.  I mentioned my background’s in history, so I actually worked in libraries for years before I actually ever got a library degree.  I kind of did it in reverse order.

MU:  So you learned it on the job.

JC:  By the time I did that, I pretty much knew quite a bit about IRs, probably more than most of the faculty did, just because they’re not working with the IR on that level.  I mean, they’re certainly familiar with the IR;  they’ve used the IR.  In fact, they publish a journal through our IR.  But in terms of the nuts and bolts of running one, they wouldn’t have had that experience.  That’s usually why I get called in to explain, that bit of it.

MU:  It sounds like a fascinating challenge, and I think much needed.  It strikes me that repositories are a fundamental part of library activity and yet, as you say, few people know much about them, and today, as much as when they started, people have to learn about them for themselves.  Let’s hope your book plugs that gap a bit and provides a much-needed introduction to the practice for people coming new to it.  I look forward to seeing what you come up with next, Josh!  Thanks very much for your time.

JC:  Sure.  Thanks for having me.