 All right. Good morning, everyone. I think we'll go ahead and get started. All right, welcome to the second day of CNI. Welcome to this session. My name is Win Coles, and I am joined today by my colleagues in the data curation network, Jake, Lisa, and Mikayla, to talk to you about what the data curation network does and the structures and tools that we have used that have helped us thrive as a network. So I'll warn you now we're going to be a little interactive. One of our goals for this talk is not just to sort of share what we've done, but also to have kind of a back and forth about ways in which some of the structures that have worked for us might work for you and what other kinds of structures and tools that you've used that have enabled success. All right. So the first part of the interactiveness. Raise your hand if you have worked on a collaborative project in our institutional or maybe between two units on campus or more to build infrastructure or services. Okay, you're in the right place. All right, tougher question. How many of you have worked on a project, a collaborative project that failed as a tough term? How about didn't live up to its potential, right? You didn't get out of it what you thought you might have gotten out of it at the start of a project. Yeah, okay. So we're going to start with now the digital interactive part of the process and ask everybody how often does the speaker say this, please get out a device and navigate here ideally. Avoid email if you can. So you can use the QR code, it'll take you straight there or if you don't want to use the QR code you can just navigate to menti.com and enter the number. So it's 31048219. I'm going to start us off with an open-ended question. Once you've entered your information, keep it handy. We're going to return to it at the end of the presentation. All right, so the question is in a short phrase or in a summary if you wanted to share the reason or reasons one or two that that project didn't live up to its potential. What are some reasons why you felt like the project didn't go the way that you had wanted it to go? All right, so while you are entering that information never been kind of get started with it. I did mention it's going to, you're going to share this right with everybody. All right, so I'm going to let that go. Please keep entering things, but in interest of time I'm now going to turn things over to Michaela. I can actually. All right, yeah, thanks, Wind. So I'm Michaela, I work at the data curation network based at the University of Minnesota. And while you're entering in some of those reasons that maybe your project didn't thrive in the way you anticipated, we're going to talk a little bit about the background of the data curation network. You'll probably hear us call it the DCN because that's just how we refer to it. So in advance that's my warning to you about acronyms. So the DCN is a partnership of 17 members now, 15 academic repositories and two nonprofit repositories that we're a community of data stewards and practitioners who collaborate to advance the data curation profession. And we do this by collaboratively curating data sets, which I'll talk about more in a minute, but we also offer educational opportunities. We create resources and best practices to help other data curators. And when possible, and in whatever venue we can, we advocate for the role of data curators and information professionals in the research data landscape. But one of the things that we might be most known for, or I think what's really interesting is what we have is a shared inter-institutional curation workflow. So I'm going to show you what that's like, but we're not going to spend too much time here. So we've got this cool little graphic that shows it. So on the left hand side, you can see the researcher deposits the data to their repository like normal. And that local curation team is still responsible for reviewing, accepting, preserving, and making accessible the data set. But when they need support, reviewing data for either a format that might be outside of their expertise or a discipline that might be outside of their expertise, that's when they tap into the DCN, which is the entire right-hand side of the graphic. So I'm not going to go into more detail on that, but what we really like to highlight is that the DCN is the human layer in your repository. We are the human in the loop. So even though our members leverage various repository technologies that are either fully vended or homegrown or open source or some kind of combination, the DCN integrates seamlessly because we can provide the necessary knowledge and expertise while accounting for local nuance. I'm going to talk a little more about the DCN history. We've been active since 2016 thanks to a planning grant and subsequent implementation grant from the Alfred P. Sloan Foundation. But since that time, we've accomplished numerous things and some of our successes include that shared curation workflow I just highlighted, those data curation primers, which are concise, iterative guides meant to walk information professionals through how to curate different formats or disciplines that they, that way they don't have to like necessarily become an expert in something they can just get a quick crash course on how do I curate this Microsoft access database. We also offer hands-on training in the form of a two-day workshop that has been incredibly well received. And we have a very active community of practice, which is one of my favorite things of the DCN. Now we have many more successes that we could talk about in more detail, but they will be published in a forthcoming report published by the University in Michigan Press that I'll talk about more in just one second. So put a pin in that. This is an overview of kind of the DCN timeline, but you'll notice on the far right-hand side at that transition to membership model. So we are fully member-funded as an organization. And as we were nearing the end of our three-year implementation grant, we thought it would be useful to conduct a project retrospective to better understand what worked, what challenges did we either overcome or maybe still grappling with. And importantly for this talk, what are the kind of structures of success? So I won't spend too much time going into detail on this, but over the course of two and a half days through guided discussions, independent reflection, and group exercises, we really took the time to understand the work of the network. And one of the things that I think will be really useful is in this forthcoming report, we're hopeful that other communities or collaborative projects interested in getting going and really maximizing the impact of their work can take the report, understand what worked for us, and maybe use it in their local application. So again, we talked about the things that we did well, the things we probably would do differently if we were starting today, but really importantly, what were the key structures that enabled our success and what were the things that we needed to keep in place post-grant in order to be successful? And with that, I turn it over to my colleague, Jake Carlson. Thanks, Makayla. So I wanna talk a little bit about the structures that we use as part of the DCN and its development, but before we get to that, I wanna talk a little bit about radical interdependence. This is the concept or the model that we use in forming the DCN. And really what we mean by radical interdependence is depending on each other is outside of our organization to do the work, right? So if Princeton sends me a dataset to curate, and I don't follow through with that, Princeton is in pretty big trouble, right? That's a pretty big situation. So it's not just working together, but if one of us is not able to do the work or fails to follow through, it really negatively impacts the other institution. But on the flip side, we can do so much more together with the skills and the knowledge and expertise that we all have that really it's worth taking that risk if we can get that model of radical interdependence correctly. So really it's about trust and vulnerability and accountability. And I don't have any ability to hold Princeton accountable or them, us in sort of an organizational standpoint, but being there because we wanna be there because we believe in the mission of the DCN and we believe that we can do more together and depending on each other in the way that we do. So pulling the resources, our knowledge to not just help us do the work that we do, but help advocate for data curation at a community level, right? To bring us together to have a real strong influence on the world that we live in. So the structures that we have really are designed to help us do that work, to enable radical interdependence. And we sort of define them in terms of administrative structures, tool-based structures and trust-based structures. So to start with administrative structures, these are the structures that enable us to do that work, to come together and to enable the data curation network to happen. And some examples of this, Makayla mentioned our grant funding. So we had a planning grant to start with, an implementation grant to move to implementing what we came up with and then the implementation of a membership-based model for the DCN. All of that really depended upon our coming together to agree what we would be doing and how we would hold each other accountable for that work. The shared governance model that we have now really is an output of working together and to see what worked in coming together and what we needed to have in place to make all of this work. We have a full-time project personnel. Makayla, our director, who keeps us running and really is the nerve center of the DCN that's been really critical for us in organizing and doing all the work. And then a consistent curator onboarding program. So we're creating a culture of practice with the DCN. When we have new members join us and come on board, we need to ensure that their understanding of what it is that we expect from them and what it is that they in turn can expect from us. We also developed a number of tool-based structures. So the administrative structures are the how, the tools are kind of the what, how do we actually do the curation work that we do? That comes from a number of different places. Collaborative office tools, that sounds kind of basic and straightforward, but how do we ensure that we're all able to access the tools that we have and the work that we need to do? So the collaboration part is really important because we're at 17 different institutions. We have a workflow management system. We use JIRA to do that. We had to really develop and think about how can every one of the DCN members access JIRA and to use that tool as easily and efficiently as they could? Our educational materials, so the work that we do, the curated workflow that we have, how do we teach that? Not just to our curators, but to the larger data curation community. The DCN primers have come out of those workshops. These are 15-page or so sort of central reference documents. If you're working with an unfamiliar format or in a new discipline, you've got a reference document to follow to ask some questions about curating that particular data set. And those have been really critical, I think, in advancing our capabilities in different types of formats and different types of disciplines. Then finally, the DCN website. So the tools that we create and the things that we do, we wanna be as open and as transparent as we possibly can with those and to make those as widely available, both within membership but outside of membership as well. And then finally, our trust-based structures. So the DCN will live and die based on trust of its membership. If we lose that trust and trust is a very fragile thing, we will not be able to succeed as the DCN. And so we have a number of trust-based structures in place. We have a yearly all-hands meeting where we get together, we engage with each other, we celebrate our success, we think about what has changed, what do we need to know, if there are trainings that we need to do to keep up, we do them there. And we start thinking about what's coming next, what do we need to do, where do we invest our time and energy in moving things forward? We have a number of special interest groups and these are groups of folks who, if there's a curator who says, I have an idea or there's an area that I'm really struggling with and there are others who are like, oh yes, we're also struggling with that or that sounds like a great idea. Can we form different interest groups to do some work and to spend some time in developing solutions or thinking about where things might need to go next? These can be anything from just discussion-based interest groups all the way to particular projects with defined outcomes. We have our code of conduct, which is similar to many code of conducts that you've seen, but really it defines our culture of practice. How do we want to work together? What things are important to us and in keeping that trust going forward? And then we have developed our mission and vision and value statements collaboratively. So we might have a group come together to draft those, but it's vetted throughout membership, not just the administrators or the institutional leads, but all membership all curators have a chance to weigh in and to respond to the work that we're doing and the documents we're creating. And then generally a sense of shared leadership and collective ownership, right? So Michaela is our director, but she does not own the DCN. I am the institutional representative for University of Michigan, but I do not own the DCN. I'm expected to work with my curators locally at Michigan and to share and bring forth their few points and their thoughts and opinions on the DCN and driving it forward. And so that was a brief discussion of our administrative and tool-based and trust structures. I'm gonna turn it back over to Lisa to continue the discussion about structures more generally. Hi everyone, Lisa Johnston here. We wanted to, again, return back to the questions that we started with at the beginning and dive a little deeper into what structures work for some of these complex collaborations. And I think what I'd like to say before we dive back into that is that these collaborations are complex. They are sometimes difficult to navigate. So I think what you've seen in some of the structures that we identified as working well for our group are really people structures. They're structures that help people connect, help people stay informed about their responsibilities, their role of how it fits into the larger collaboration. And it really helps us kind of all overcome some of the common barriers that I think we saw earlier in our list of some of the issues that really face a lot of our collaborative structures where there's maybe unclear expectations, unclear responsibilities for different group or team members. There's competing priorities. There's not really a very clear mission statement. Let's see, oh, these are popping up as we go. Different goals from different partners. We actually embraced that. Some of the work that we did early on was really diving deep into what do you need as an institution as a part of this larger project? And let's build that into our mission and make sure that we check in every year with each one of our members to ensure that we're still reinforcing and reflecting that local need and that local priority. One of the other things we really tried to do to kind of not get in the way of what does happen locally when it comes to data curation. I think it could have been really easy for us to just come around and say, well, we know how to do it well and we're just gonna pick up the slack and do your data curation and work directly with your researchers. And that would really break, again, that people connection locally. So we don't wanna interject ourselves in that service that is really there at that local campus and not allow that strength in relationship to happen. So I think we could go to the next slide here. Do I know how to do that? Oh, good. So with some of those examples in mind, we'd like you to return back to thementi.com. The code is at the top. And we'd love to hear about which structures that you heard that resonate with you that are things that you've seen used in the past or you've personally used in the past with your collaborative projects. And tell us what stood out or what did something maybe you would agree with. And we just kind of listed them out here on the left in case you forgot. All right, so we can see some of the votes in the trust and administrative areas, that's great. I'm really excited to learn about what this something else might be, that's great. All right. And then I think I've got one more question I'll jump to if I see any more dots coming our way. And then finally, if you could list out maybe what are those additional things or what were we missing that you, oh, I didn't see that one go. That you would like to add to the conversation. And we can also take examples through the microphone or other perspectives that you would like to bring to this idea of collaborative and enabling structures that ensure our cross institutional and our collaborative projects can be successful. So I think I'll just switch to microphone that last minute to use on microphone. So what were the other things? You can shout them out, I'll repeat them. One of the important factors I think is having the people who are in charge remain a stable group. The more it changes the more the direction can change and that can be a problem. That resonates with me as someone who transitioned on the group. Our challenges have been encountered recently, right? We started with a core group of six. Lisa was our first director. And in a way it was sort of like a startup situation, right? We had a lot of ideas, had a lot of things that we wanted to try out, really went far and did a lot of those initial ideas. And then our initial program director left and took a different job. And so we were at the point in the startup of are we gonna be able to come together and keep things going the way that it had been? Or like so many startups, are we gonna have to fold? I think we've done, I think a pretty good job under McKayla as our new director of keeping things going and keeping our values together and keeping our structures there. But there are still some questions about what is the future of the DCN and how will we evolve and move forward given that our leadership has changed. And I think I would also add the strength of the governance model, the role of having a document that you can go back to that really articulates how you make decisions as a group, how you vote on certain things. We actually have a governance board that meets monthly to include representatives from each of the institutions that make decisions and help drive the mission of the work and the project forward. Hi, thanks. One of the things that resonated with me is how you seem to have created this sort of lightweight layer on top of these repositories and the local kind of situation that's happening at each institution and then providing expertise on top of that or a place to turn to for additional expertise but keeping it fairly lightweight. Yeah, thank you. I think there was maybe an allure of trying to go down kind of a shared technology path early on in early stages and I think we recognized that the problems of disagreement and the problems of not having a lot of control over how those decisions are made at each of our institutions would really prevent us from moving forward. So focusing on that human layer was really essential and has paid off in ways that we just did not expect. We've really gone down new paths and found areas of community need that we're really able to address and build on. Any other discussion points? So one of the things that seems particularly exciting is you have a network of how many expert data curators now, individual? 60. Which is a big number. So we are all, I think, quite excited by the advent of the OSTP memo from August and every campus I've talked to is trying to figure this out individually. You have 60 people who are talking to each other while each of their campuses tries to figure it out. Do you see DCN helping either shed light or forge progress on that front? Yeah, that's absolutely. We've been having those same conversations as well. I think one of the ideas that really resonated with the group is that we could provide a space for repositories to come together and really assess their readiness for meeting not just the Nelson memo expectations for data sharing, but also the desirable characteristics of a data repository. So those were earlier guidelines that came out right around the time NIH released their data sharing requirements. So that memo of 12 different characteristics, we've done self assessments within our own repositories of how we meet them. We don't meet them all. And so really building out a roadmap for how we can establish additional services and additional workflows and additional technology tooling to really demonstrate that we are providing a trusted and reliable source for our researchers to be sharing their data. Yeah. So exclamation point at the end of that one. I think one of the reasons we really were intrigued by the notion of a data curation network in the first place was we need to know how to do this work. There were not a whole lot of models out there for doing data curation work, particularly in libraries, but even just generally, right? So coming together and figuring out collectively made a lot more sense than trying to figure it out at Michigan, at Princeton, at Minnesota separately, right? We have more power coming together. What I hear now with the OSTP memo with other initiatives that are here is how do we do this on a much grander scale, right? And not just in libraries, but in institutions and in funding agencies. And my hope is that we've done enough to figure out how we do it within both locally as well as across a network. We really have learned a lot from each other. My local program is much, much better as a result of having been a part of the DCN. And so my hope is that with the expertise and our experiences that we've had, we can start to influence the larger culture of practice around data curation across the United States. It was great to see these slides and to understand more about what DCN has accomplished. One of the questions I'm thinking about on my campus at Stanford is the role of data curators in helping to spread best practices or awareness of software tools, even basic things like open refine or learning Python, to people in biology labs, to people in English departments who are now having to curate data. And I loved how you guys talked about how you as data curators could look at a data set, make recommendations, maybe pass it onto the network, but eventually give it back to the patron. And the question is how can we advance expertise, competence around data curation at the patron level without taking all of that on ourselves, which just doesn't scale. So as somebody who's trying to hire more data curators and really wanting them to be patron facing, but not having them do the academic's work for them, I was curious how DCN is thinking about that kind of education and skill sharing problem. DCN, we do have our education programs, but I'll admit that those are more geared at the moment towards data curators. But some of the things we've been thinking about in the education committee, and I can't see if Cynthia Hudson Vitale is in the room, but she is also someone to connect with about DCN education. We're thinking more about how do we take our extant resources and workshops and translate them to a researcher perspective. So that way when that data curator is out on the call, they can then just say like, hey, by the way, here's this data primer that you might wanna familiarize yourself with that's not written for a curator perspective. Yeah, Wind. Well, I was gonna say too, even the primers, I know of curators who have used them as teaching tools, right? So they weren't intended to be researcher facing, but in being accessible to curators, they are accessible to researchers too as a teaching tool, right? So you've got this kind of data, here are the things that you should be thinking about at the beginning of the project that are gonna make it easier later, all that kind of thing. I'll just add one more thing and maybe it's a little bit provocative, but I think our goal is not to curate. You know, what we want is really well managed and well described data at the beginning. So yes, we take that opportunity as a teaching moment when we are working with researchers and publishing their data to talk through about some best practices as well. If you started to read me file as you were going, we could speed this along. And we're also doing a lot of training and workshops in those departments. So the role of a curator, I think it really helps inform us of where those issues lie and we're in that immediate moment of, yeah, maybe I didn't do it perfectly, but I don't care, I need to get this done. But that helps inform our training to go back and to show those examples and to say, here's how we can improve at the start. So that when it comes in, you know, we're doing, you know, we're not doing a lot of extra work in building out all those documentation elements that could have been there from the start. I think Lisa's right. Like we certainly see a lot of different data sets in a lot of different states. Our goal really is to get data sets that are submitted in a high enough quality state that they really do add value to the community that they're trying to connect with. And to have that, I mean, the researcher at the end of the day owns that particular process. It's their data set. It's their name on the data set. I'm trying to develop a data set that they'd be proud to have their name attached to. But ultimately, it's their responsibility to do the work and to make that something that, you know, is useful and valuable to their particular community. So we ask a lot of questions, but the answers will depend upon the nature of the researcher and what they want to invest, what's expected by the community, what they would like to receive. And so I think our goal really is to raise awareness of those kinds of factors and elements and to, you know, hopefully see them through, again, with the researcher at the core of that particular process. Yeah, we are at time. So if you have follow-up questions, please come find us afterwards. I'll turn it over to the pod team. Thank you.