 Thank you for coming to our project briefing on expanding research data services. Today you're going to hear two academic library perspectives about expanding research data services. First up, I'm Bryan Sinclair from Georgia State University Library. So coming up first will be Mandy Swigart-Hobo and I talking about research data services at Georgia State. And then Michelle Claiborne will follow us speaking about the Director of Research Data Services at University of Virginia Libraries and we're talking about how they are expanding and working with campus networks to even expand research data services even further. So we'll go ahead and get started. Our portion is more than data management plans exploring new outreach opportunities to expanded research data services. Mandy is passing out a handout that we have been distributing on campus. She'll tell you more about it but some of our services and personnel have expanded since that handout was created. So first and foremost or when we talk about expanding research data services at least in our culture in our context we're talking about data services and we'll define that or get more into the details for all campus users. So the undergrad, the grad student, the postdoc, the staff member, the senior faculty member, we're talking about services for all of those audiences all of those potential users. We are interested in understanding campus needs. This involves often running ideas by our Student Library Advisory Council. This is their last meeting last week. It's a smaller group because it was during finals. We run ideas by them. That's our Dean of Libraries and our Assessment Library and Jennifer Jones. We offer multiple focus groups and feedback stations in the library and most important for me is we do face-to-face surveys man on the street or a woman on the street interviews with students outside the library to find out what their needs are. So that kind of sets the stage. So for example in spring 2015 we conducted we the library conducted a student technology needs survey and we did not lead the witnesses in this. We just said what are your needs? What are your technology needs? What are your training needs student? What are your resource needs? And the number one response or the number one need was Excel and that was that we found that very interesting. We also surveyed our faculty and said what's the number one training need that your students need and they said Excel and they also mentioned several statistical software analysis tools like STATA, SPSS, etc. But we did ask ourselves what's the library's role here and what the campus need is there. What's the library's role? So we also are tuned to the research needs of our faculty, not just that undergraduate student who needs help with Excel and supporting our research mission. We are a fairly new up-and-coming research university. Research funding has grown 100% of the last five years. It's over $120 million a year and growing. It's on a trajectory. So that's important too. We have to keep, as we all do, be mindful of our other researchers. So in the 2013, 2012-2013 school year, we did establish a data management advisory team. And that was the big year, 2013, if you remember, the Office of Science and Technology Policy issued its memorandum that any federal agency with over $100 million in annual expenditures is to develop a plan to increase publicly funded research data journal articles as well. There was an executive order from President Obama. There were OMB memos to the federal agencies about open data. So we embrace that as many of you who work in libraries may have with services. We specifically promoted the DMP tool. We work through our office of our research office, our VP for research. And he's very, very, very good meetings with the VP for research. And we had some good business in that area. We talked about a dozen PIs, mostly in the social sciences and humanities, and that continues to today, geosciences, education, anthropology, etc. And but I will say that business is not as robust as it was three or four years ago. It is on the decline and we're not sure why. So but it is a service we provide. Now, in the last year, in this year, 2016, we have a new Dean, Jeff Steeley, who came to us, and we have been involved in library strategic planning. And expanding research data services has been identified as a strategic priority for our library. So there are many diagrams illustrating the research data lifecycle or the research lifecycle. And this is ours. It's a little bit smaller than most, right? And and many chair the team that came up with this. And the ideas I'm going to be talking about are really Mandy's. So academic libraries in general, are well known for supporting it and the exploring and questioning phase. That's the first one with binoculars. So we're and that's what we what we have been describing recently as the little R of research. That's undergraduate research papers assisting with lit reviews, etc. And many of us like like you and like us, perhaps like you have been working with the sharing and documenting through data management plans, data storage, etc. etc. But but we also see a need to support the big R phases of the research cycle. And in which all levels of researchers undergrad grad faculty are designing and planning their research project are it's finding existing data, exploring appropriate data analysis tools where they're analyzing their data and creating their research outputs, cleaning up messy data, performing data analysis, creating data visualizations. There's been a real gap on our campus in support of these phases, the ones on the right. And the library seeks to fill that gap. So as we talk about expanding research data services, it should be pointed out, as I mentioned, we are a younger research university. We were in 1995 designated as a research institution in the state of Georgia by our by our university system. But we don't because of that, we don't have nearly the political battles, turf wars, and institutional culture to fight. So if the library wants to recruit higher and expand services and GIS, or in social sciences data or statistics or even survey design, which is increasing the we don't normally run into a lot of turf wars or battles. There's a lot of teaching learning research support to go around. And most people welcome the library getting into this business. So here's a digital sign in the library. That's advertising some of our many workshops that we did this last semester. And you can see, these aren't your typical library workshops, maybe they are in your library. Statistics in the real world, again, Excel pivot tables was a key element of that we were told this is something we want to know more about. I'm using open refine to clean your data, finding data sets in ICPSR, for mapping workshops, both getting into the weeds with GIS and then also just how to create a map for your for your research paper. And so again, what what may set us apart is that we strive to collaborate with all researchers, as I've said, including the undergrad, who are typically underserved in the growing and critical areas of data science, we believe that data literacy goes hand in hand with information literacy and digital literacy, and that there are training and consultation needs we've only begun to tap into. And this is one of our GRAs actually, giving one of her popular Tableau workshops in the library recently. So with that, I'll turn it over to Mandy, who will continue talking about Georgia State. All right. Hello, everybody. So I am the team leader or interim team leader, because I'm trying it out for a year deciding if it's going to give me a, you know, heartburn, I already have heartburn, more heartburn than I already have. But interim team leader for research data services. So let me make sure I know how to operate this. Okay, so these are general areas of data services. And some of you who came in a later might not have a flyer. But those of you who did get in here sooner, and when we still have flyers to pass out, we have this flyer that gives kind of the details of in these general areas what types of specific tools or services we provide. And actually, there's also a web address on there that if you go to our web address, you'll get more details about this. And actually, since this flyer was printed, we've recently added a digital scholarship librarian, who he has skills that overlap with data services as well. So I immediately harangued him into putting his initials on these and saying he would provide services. I didn't give him much time. So who is providing these services? Well, we have librarians providing these services. And then we also have grad students providing some of the services. And for the librarians, this is a great opportunity because it's helping us expand our roles and kind grow professionally. And for the grads, it's a great opportunity for them, because it is going to make them more marketable, to be able to say that they've done these things. How we're providing our services, we offer drop in help, as well as scheduled consults. And then we have some open workshops, like the ones that Brian showed that were advertised on those digital signs. But we also do custom workshops for people. If they need those, we work with classes custom, do custom sessions for them. So an example from this semester that it was a kind of a nice organic way of helping, I use the word organic. Organic way of helping people was we were doing drop in assistance hours for SPSS help. And we had several undergraduate students from the same statistics class and sociology coming to ask for assistance. And we were able to figure out who the instructor was, which it was a graduate instructor. And I'm also the sociology librarian. So I knew her. And she's my Facebook friend. And I said, hey, Stephanie, we get a lot of students in here needing help with SPSS wondering, do you need a classroom with a computer lab with computers that you could come in and we could float around in troubleshoot with you of helping them do that. And she said, that would be great. And we ended up doing four sessions that way. So it was a nice way to illustrate that we really try to work for custom in on demand needs of our users. Generally, our website, these are some snapshots of our website. So we have our people who are providing the services, they're little initial bubbles that our graphic designer in the library made for us. But if you click on those, you will automatically send an email to somebody. We list again, all our different services that we provide. And again, we have the little initial bubbles so people know exactly who they can contact. If they click on that, they can send a direct email to those people. We included a little feedback box on there that now it's kind of hard to see from here. But people can give feedback on the workshops, give feedback on individual consultations, or just share ideas with us in general of what they want to see for future data services. And then we have a calendar that lists exactly when our drop in hours are and when workshops are that you can filter to if you're looking specifically for somebody's specific drop in hours, you can find them and so on. You the flyer that you have, again, we had we developed this flyer that was distributed by PDF to all the deans on campus. I think our library Dean, Jeff Steeley sent it to them. And it trickled down through the department. I know for a fact that the sociology department chair sent it to his grad listserv and his faculty listserv because I'm on those. So I'm interloping and know that it made its way down. We also provide a PDF of this to another entity on campus that's through our IT department that does provide some training workshops on things like Excel and SPSS and we are working closely with them so we don't necessarily duplicate each other but that we compliment each other. And also they don't have the capacity to provide one on one consultations. So they really depend on and refer to us to say, you know, oh, you need a one on one one on one consultation. Go see this person contact this person. We also asked our liaison librarians to forward information about services to various departments, including PDF of the flyer. But we also combed the class schedule we're doing this again for next semester, comb the schedule looking for specific classes to contact those instructors to do targeted marketing towards them regarding specific workshops or services. So for example, for the statistics in the real world workshop, I contacted all the statistics instructors on campus and said, hey, this might be something your students are interested in. Please consider letting them know about it and I know that some people did it before that on. That's me on the screen. And a couple of weeks ago, literally a couple weeks ago, we gave an overview of our services to the deans of the various colleges and schools. I'm asking them to actively promote this services again. I mean, they had sent the PDF flyer at the end of the semester, but it's always good to keep telling people about our services. But to also ask them, you know, what what can you tell us? What input can you give us to help us improve our services? At this meeting, I also distributed a handout that's printed on the back of your, your handout. So still cited. And on the back, we distributed it to the deans to give them a sense of the great work we've been doing so far. Again, we really just rolled this program out in July or for this semester. So even so far, you know, we gave them some of our numbers about our consults, but we gave them some good stories, some little snippets of how we've helped people. So I shared the fact about that SPSS class that we we held because of recognizing a direct need for that. And in terms of assessment, in addition to evaluation of workshops and individual consults, etc, where every year, roughly, we're hoping to try to collect at least like 10 good stories or success stories could have more like anecdotal or kind of testimonials in some way. So this handout is kind of a good start towards that. Recently, our, I think this is the research computing team, which is through the Research University Research Services Administration or the Research Office on campus. They are have included a link to our webpage for help on data services support. And they they're very supportive of us. I know like Brian said, we don't tend to have turf wars. I'm going to be a little more cynical and say we don't have turf wars because they don't want to do it. Or they don't want to do it. Or they don't have the resources to do it. So they're perfectly happy to be like, Oh, look, the library will do all these things for you. Get out of our face. So that's that's kind of how I see it more. But I probably shouldn't put that on the Q&A. So I'm going to turn it back over to Brian. Okay, I can kind of wrap us up. We're going to wrap it up real quick, because Michelle's we want to get to Michelle. And if you could just save your questions for after Michelle's presentation, we hope to leave some time to have a dialogue with you about some of this because you probably have questions. So what's next for us? And this I need to I need to hit a bunch of points real fast because I need to get turned over to Michelle. And oh, and by the way, when we get to the Q&A time at the end, that's not recorded. So we can have some real discussion here. You can be cynical. I'm very positive. I'm very optimistic. And if you want to be cynical, that's good too. So anyway, so what's next? Let me talk about our reachable, the low hanging through the reachables for us right now. This is stuff we're actually doing. So we're going to do more focus groups with faculty and grad students, targeting faculty, specifically, as Mandy said, you want combing those, those catalogs and finding out where these classes are and targeting faculty more specifically. Continued partnerships with our research computing and the VP for Research of Office. Now there's two types of data. We haven't started data storage yet. That's a whole other presentation. But I should say there's two types of data storage we work with primarily that's your working data storage, the collaborative space where you can do your work with your fellow researchers, colleagues, grad students, students, and then there's the long term data preservation, right? So I would say most of our questions that I've anecdotally, I don't have any data have been more than working, I need a collaborative space to store my data and work with my team. And we are working with research solutions, they have identified and someone in this room may know more about this Dropbox for Business, as a possible thing that the library might help with to support faculty. We are told it's HIPAA, FERPA, etc. compliant. So Dropbox for Business may be our next kind of foray into this. And of course OSF, things like that, for long term preservation, we have already had meetings with research computing, that is a big, big push from the library. But we must work with our research computing office on some of those initiatives. But they're in the works. Harder to reach, but there is commitment from, I'm administration, so there's commitment, Mandy, there's commitment. So I'm not, I like the 10 good stories that Mandy talked about, I'm a more anecdotal, qualitative person, but we need a quantitative expert. We need to hire a quantitative expert in our library, specifically for statistical consulting and data cleaning. That is a big, when we talk about data management, faculty say, oh, you can help me clean my data. No, we're not using the same language, but we need someone who's a data cleaning person. More expertise in survey design, specifically in use of Qualtrics. And there's commitment there, increased staffing and GIS support, specifically someone who can get more in depth with our entry license and all those moving pieces if you're familiar with that. Secure physical and virtual spaces for working with restricted data. We have this great space in our library called Curve. It's very open. But if you're dealing with restricted use data, or more need to be more private, it's terrible place. So we need to find a place for those researchers to work with that type of data set. And then consultation and writing things other than D data management plans. So, for example, an RB proposal using restricted data, like I mentioned, we could help with that. That's something like we understand that some of the boilerplate. So with no further ado, save your questions, we appreciate your attention. Michelle is going to talk and then hopefully we'll have some dialogue toward the end. Awesome. So I'm Michelle Playborn. I'm the Director of Research Data Services at the University of Virginia Library. And I use this metaphor, the TARDIS, if you're not Dr. Who fans, I apologize, you're going to have a long 15 or 20 minutes as I have Dr. Who quotes for you each time. But, you know, it's kind of my current metaphor for the networks and partnerships that we've been building with our research support partners throughout the university. And with the idea that by entering in any of those ways, right, through any of those partnerships, through any of those doors, the hope is researchers gain access to a distributed set of research, expertise and resources, making us all appear kind of bigger on the inside, like Dr. Who's beloved time traveling police box. So today I want to tell you a little bit more about those networks and the initiatives they've helped produce and as well some of the challenges or barriers that we've encountered. So we just hit the, oh, that worked too. Great. So again, I want to let you just know what research data services means at our library right now. And the library established this unit that we call Research Data Services only in October of 2013, merging together sort of for some longer term and some new services that we were creating, including our oldest, our three-year-old at the time, data management consulting group who had done a lot of great early work advocating for and advising on data management planning and sharing, archiving and curation. And then we also at that time had a newly created stat lab, which provides consultations, training and support for data analysis, visualization, wrangling and statistical computation. We also had a newly hired data library and providing data discovery, expertise and working with researchers to locate data and to build the library's data collections. And at this point we also adopted in some ways our research support software support group that had formerly been housed in the Information Technology main IT center at UVA. So they became part of the library and their role is really about helping researchers locate access and install the site license software research software titles available to them. So this kind of all came together and we became Research Data Services. So our work in the first three years has focused on a few activities across all these services including consultations and collaborations with researchers, education and training for data intensive research approaches and I'm going to briefly talk about some of those and then get on to the other one which is really about the building of partnerships and programs within the university to promote data, computational and quantitative research across our campus. So in the summer of 2016 we had existed for three years and I thought that was a great time to pull all our data about what we've been doing and send it up the train up the line right and so the advent of RDS as a combined unit I think really helped situate the library more centrally in the research process particularly of our quantitatively and computationally oriented scholars. So one of our key one of our primary activities is this direct engagement with researchers through our consultations and collaborations and so in our first three years we hosted nearly 1,800 consultations with researchers. The majority of those actually did center around data analysis, wrangling and statistics about 60% of them. So I checked I last night I looked at our our forums and we have another 375 records from this fall that I that aren't represented here and I haven't broken them down yet but I was like pretty excited about that actually. Our consultations I'm particularly excited that our consultations really reach across the entire university. So through the summer 2016 we've met with about 800 different researchers because we because about 45% of our 45% of our researchers meet with that meet with multiple of us or meet with us repeatedly right which in my sign is it I mean I think that's a really great sign that they're coming back or they're cross referencing we're able to kind of bring something. Most people come to the stat lab and we're able to say hey you know you should talk to our data discovery person or hey you should talk to our data management folks right and it's able to cross reference and people come back and I think that's a marker of the value they ascribe to the work we do. But also we've reached all 11 schools I got off track all 11 school like we have 11 schools at UVA and we've worked with people and all of them right and I think that again makes me really proud of the work the team does that we've managed to kind of touch at some level some more than others the College of Arts Sciences which is at top one is the biggest one that's all of our humanities social sciences and sciences so that weighs everything but that we make connections across the campus and we worked with researchers at all levels right from novice undergraduate researchers to expert faculty researchers or sometimes novice faculty researchers when they're entering a new domain right we've kind of run the gamut in that way the bulk of our engagement that was with graduate students in about 45% of the folks that we worked with are grad students which is a community we explicitly targeted early on so in these kind of collaborations or consultations we do all sorts of things right as represented by represented by our service we in just to give a few illustrations and I need to put these together in a love I love Mandy Slyer so I'll work on that we worked with the business faculty to reshape his experimental data and implement a multi-level model to estimate treatment effects for a repeated measures design right something he was told to do by reviewer but didn't quite know how to do right we worked with an interdisciplinary team of faculty to draft a data management plan for an NSF sustainability research network proposal which was more challenging the most because it was so interdisciplinary nobody wanted to you know that wasn't just one faculty it was like oh there's ten who's gonna take that you know that potato right and so that one was when I think we did a lot more work than we might normally have done we consulted repeatedly with the politics faculty member who was trying to acquire some key international time series data that he needed to finish his book right and we were able to help him get that and we worked with our office of institutional assessment to help automate their survey reports by helping them write reproducible scripts and make their work more efficient and so we kind of again that's just to say we over a variety of services and a variety of different kinds of users we've interacted with the second kind of big effort we've made is in our workshops in our training program so we you know we're a library and education is a core mission so we've been developing our workshop workshop series and to me what we see is sort of a really the widest widespread knowledge needs right we've tried to hit the things that many many people have been asking about and so but particularly around data and computational skills and particularly the kinds of the hidden knowledge there's things that people think you know but nobody teaches and it's not part of any formal curriculum that's really the year for us the sweet spot so we don't intend these to be substitutes for departmental curricula as something I had to explain to some faculty in some departments a couple of times right but in fact to be to compliment those things by providing a pool of collective resources from research where researchers in any field could draw from so in the first three years we've offered 90 workshops open to the whole community including people outside the community who sometimes come for which we had nearly 1900 learners registered the workshop offerings again range from introduction to common analysis environments introductions to R visualizing it was ggplotinr or stata or sss or sass all the way up to sort of advanced statistical methods so we offer things like multiple imputation for missing data or machine learning right introductions to these things as well as well as computation is sort of programming approaches we work we have workshops on python interfacing with api's getting a custom to the command line and things like that oh that was an ending clause when I should have not ended yet as well as research to management workshops where we do data management planning as well as things like introduction to database design and sort of getting people familiar with other sorts of tools so we've been working to build kind of new efficiencies as well by recruiting experts from across our campus and providing them a ready made platform to share their knowledge why I go around anytime I need somebody I'm like you know cool stuff I say hey would you like to do a workshop on that and it's surprising how often they say yes right because all they have to do is prepare the thing and show up and I've got this whole system right that they can just walk into and increasingly I realize that's actually a valuable resource in itself right even if the experts are in the library having kind of a thing that they don't have to do any logistics for makes them so much more willing to come and share their knowledge with our camp with our users and our campus oh and the word cloud just so you know I should probably say what that is that's a word cloud of all our workshop tiles with the size tied to the number of registered users registrants for their court for the workshop so you can see there are a big thing for us but these workshops to have proven a university wide resource and they've drawn interest from across campus with again researchers coming from all 11 schools and several additional centers again graduate students are a modal user about 60% who come to the workshops or graduate students but it's across it runs the gamut from faculty to undergrads to research staff to other staff so this is kind of what we spent a lot of time on in our first three years as well as partnerships but there was a lot going on at the university that really catalyzed some of the partnership work we wanted to do so there's a lot changing on campus all around the same time and here I'm going to highlight just a few of those things that really altered our context in ways that we were able to leverage one was that UVA established data science institutes this is the first of several planned pan university institutes and the DSI right we call it was really there to kind of be a hub of data science activity on on campus and they also provide a master's program a master's in data science program we also hired a new VP for IT with an emphasis on academic computing so our IT central IT department had become increasingly focused on enterprise systems something then Cliff kind of referenced in the opening remarks as well and we've we've been hearing for a while from a lot of researchers that they felt like the university at large had been neglecting the needs of more academic computing the things that weren't generalized and universal for everybody that needed special systems and so his one of his charges was to do that so we suddenly had someone at a high level with interest in the university the school of medicine and the health system is a big part of our university the school of medicine just hired their first director of research computing which sort of marked a growing interest in investment on their part which again if you have a medical school in the health system like that's a big influencer of the university at large and then uva also spent about five million dollars across two different high-end computing platforms we purchased a couple years ago a bunch better hbc cluster that we had previously had called revanna and then more recently uh something we call ivy which is a secure compute environment for protected data and so it was meant to really make the people who had to work with hippo for pedata in a compute environment not just storage but where the compute and the storage are together it was meant to meet that need and so we had new systems in place new people in place and that really provided us quite a few opportunities you know to move forward in new ways so at the onset of the data science institute the new director approached the library about bringing me on board in a partial appointment as the first associate director and my charge was data services and infrastructure which i felt like i was kind of doing anyway from the library point of view but it gave me a different platform to do it which was kind of fun and interesting although if you ever do those partial appointments you know it's not really it's like nothing else went away right you just had more but i think it was a great chance for us to build some really enduring bridges with this new institute for around which there was a lot of excitement on grounds on campus we call it grounds i'm sorry keep trying to correct that um and so we've done things like you know work with them on delivering boot camps to their students they've worked with us on resourcing workshops and boot camps that we've where we bring in outside experts like software carpentry or the center for open science and we just in fact wouldn't think i'm really excited about we've just finished a two-year pilot between that was a partnership from the library and the data science institute where the library staff offered one credit short courses because our library can't offer credit we offered one credit short courses through the data science institute so open to all learners so anyone across the university could take class where they got credit for things like data wrangling in r or text is data or applied causal inference or data wrangling in python with a four we've done so far which both helped us have a new platform because we've been working on how could we you know the main question after our workshops is how could i learn more um i've answers to that that i'll say for cuny but this wasn't one opportunity like well here's a way of doing it where you feel like it's not just extra um it's not just an add-on and it also helped meet the need of the data science institute who has spent most of its curricular time on developing their master's program which isn't open to anybody besides their students right and so people are like hey what are you doing for us another opportunity between all while this stuff is changing was created again by the new vp it so our unit in the library had been working to build relationships with colleagues in a complementary unit called advanced research computing services which is the the computational experts that really support the hpc platform um we've been meeting them regularly and trying to kind of figure out how we might work better together but this new environment and particularly the new vp it gave those efforts a really big boost so the vp it kind of took on the long recognized need to bring together related research support efforts that were spread all around grounds campus to help our community discover these disparate experts so recently these networks have been formalized into cadre so this is a screenshot of our web over to the right of our web new web page um the computational and data resource exchange which is a partnership of research data well of all these for research data services our advanced research computing services the vp it's office the data science institute our central it's is research cyber infrastructure group our health sciences library and their bioconnector efforts in our school of medicine research computing so I have room for everything so there's a spell out the acronyms in person we you know we all came together and you know had 15 people at a table over many months where we jointly designed a web page which is fun but there it is and so we finally got it done and it was really exciting um and I in part of what you know the benefit is having is to answer that question that researchers say is like why can't I find it all in one place that's what we're really trying to do here the web page is linked to from all the partners websites and again the hope is that there are all sorts of portals on which you might come through to find a central place where you can find eventually I'm sure not everything yet but eventually everything that we can come up with that would be a resource particularly adventurous to computationally data-oriented researchers um so that's that actually on the one hand it's just a web page but it represents a really big success in a lot of effort and I'm really proud of it actually and the cadre effort in general we meet regularly has been a great vehicle for bringing many of the dyadic partnerships that we already had into one big unit right so we already worked with most of these people and and sort of just between ourselves and this kind of brought it all together so a key goal of this effort is to improve coordination and exposure of data and computationally oriented infrastructure training and expertise and just this after we've launched this page we've started inviting now that we have something we've started inviting new folks in to be part of the group including a new legal empirical what's called legal empirical labs something like that's the data folks in the law school and a visualization lab in the architecture school and our digital humanities efforts and from our colleagues down the hall in the library's scholars lab or I've been trying to get them kind of involved to make the network bigger so we've had a lot of new partnerships and programs that began to actually result began to develop excuse me even as a result of this right this big group of people meeting regularly and now it offered new things many of them still dyadic the things that might not have happened otherwise including we began working more closely with the vpr's office our best president for researchers office this fall on funding discovery they recently licensed some funded discovery tools they don't have the capacity to sort of help users with them and we said hey we're librarians we'll do it right you want us to have you want to help them find make a query that's going to get them good results on a weekly basis yeah I think that's probably right in their skill set and it's a great new way for us to kind of interact with our research community we've worked with the vpr's office and the center for open science on getting into the osf for institutions world we have a uv a branded osf page now and so we've been part of these groups kind of bringing these things on board in ways I'm not entirely sure probably maybe they would have happened anyway but I feel like it was all much smoother because of these efforts and there's others but I'll I'll stop there oh no I do want to say a couple more actually there's I know as I've seen this is why I wrote it down I to be honest I feel like our rds work with our own colleagues and our thriving dh community has really spun up more because of because of these broader networks and we've been able to do more to connect that vibrant work with our science and engineering groups for instance in a way that I maybe was a little bit disconnected before and having kind of sort of have feeling like you got partners right just makes things easier so our rds group sponsored the first uv a our meet up in virginia and we have a monthly meeting with like 400 400 people will come but 400 people are on the meet up page and it's really fun right and it was easier to do because we know we had people that would be would have our back on it our library rds sponsored the first the inaugural data fest at uva which is an american statistical association program that's the 72 hour data hackathon style data challenge for undergraduates in in part because we actually felt like we weren't targeting undergraduates very well and so we wanted to do something but also it was really easy to get partners and funding from about seven different partners for that event which was a really big deal and having this kind of backing having this kind of network again really facilitated those efforts so it has not been without downsides though right all of this integration and partnership has presented challenges some expected and a few that you know in retrospect seem entirely predictable but I hadn't anticipated in advance and so among the expected things that I won't spend much time on this so we can have time per conversation you know of course our time constraints right maintaining relationships with seven eight new partners it's time consuming I'm surprised at how much time it takes there's staff constraints right as we expand this new areas we have a lot more people saying hey maybe the library would like to be part of this and we want to say yes and yet we don't actually have more people and so we have to be really really vigilant about shifting work instead of just adding to it and that's you know it's you you guys know that's hard cultural collisions right not all of our partners have the same expectations or practice of sharing information or frankly of sharing credit right and so these are things we have to kind of continue to work through and that and that credit part kind of matters because of one of the unexpected challenges that I'll get to which was this that partnerships you know they diffuse responsibility we kind of that's predictable that has reasonably predictable consequences things like you don't have the same time horizons right not everybody is willing to work on things not entirely sure how easy it's going to be to sustain a webpage maintained by seven different units right I guess we'll see but what I was less sort of prepared for was the diffused credit or recognition and so while our institution regularly voices the desire for more partnerships and coordination to help leverage dispersed resources that's not in fact how our budgets work our budgets are by separate units and not everybody you know there are people of different power in these partnerships and some people have more power to advocate and claim credit and get resources for their unit than others and it's just asymmetric right and that's not something I had really planned for and it's one of those roles where I think library leadership really has to be involved precisely because of those things it's also problematic because our users don't always know where help is coming from when they come through these other systems which is fine for them but we also have a new budget model where chairs and departments are being given a list of exactly how much of their budget is being apportioned out to central services like IT and the library and ARGs and things like that and so they're seeing that and so on that point really does we really do want them to connect it to us in a way that the partnerships frankly make slightly harder and secondly the partnerships kind of obscure resource needs right but precisely because multiple partnerships the networks and collaborative programs make a team look bigger which I'm going to sell is largely a benefit it can make it harder for administrators to see the insufficiencies and support right they have a list that five people are on this team it's not necessarily clear that only five percent of any of those people's time is actually on this project and so it looks like things are about better resource than they are and maybe less benignly it gives them cover for insufficient resourcing so that one's bitten us in a way that I'll happy to elaborate on in Q&A a little bit but in general this call to be dynamic and responsive which we're all trying really hard to do makes the kind of formal institutions we normally rely on to guard against these things like MOUs and sort of written arrangements just a lot less practical right when you're trying to be nimble and so we think we really need library leadership involved in a pretty direct way that's you know in some of these conversations to guard against these things so before I end this is just a picture of the team my research data services team so I wanted to give you a sense of what we have so while RDS was developing our library also went through a reorganization in the summer of 2015 where the RDS team became part and the social natural engineering science liaison team got merged so I'm actually director of research data sciences and social natural engineering sciences so we're responsible for data services as well as liaison work with the social natural engineering science departments at UVA which is a pretty big population but one of the things this is allowed us to do is seriously cross pollinate so that most of the data service experts now liaison with the department as well which gives them kind of insight sometimes into that work and most of our library liaison experts I've been coming into our data services efforts as well helping research with data management plans helping with data discovery in particular and I think that cross pollination has been really effective we're really mixed bunch so I just kind of added backgrounds a little bit from the group so I'm going to I would argue if you push me that that mix is one of our biggest assets and I'm happy to talk more about that later but now I'm going to end by saying just a few things that are coming coming down the pipe for us because we are nimble and responsive and things continue to change right so some of the things we're working on now and I would love any kind of I conversation or feedback or suggestions about any of these things is that we're still working on integrating these teams the RDS and the social natural engineering sciences RDS to date has had no identifyable space or technologies we just have our offices and so one of the conversations that's just begun is whether we can have some sort of welcoming storefront place for our work and some new technologies we're acquiring and because I would say one of the benefits of the there are many benefits to the integration of the teams but one of them is that it's really started a conversation within our group about efforts to better support qualitative research in addition to quantitative research and that's been really driven by the social science librarians in a way that I'm not sure we would have even started down on our own and finally we're taking some first steps into our own internally developed projects most of the work we do is very responsive to the research community and we've got we're starting we're starting our data for democracy lab very excited about it but I'm not gonna it's new and I not sure there's much to tell you yet but the idea is to pull streams of information about the presidency together and extract features from it in a weekly basis so we can kind of all know what's being said in different outlets I can't imagine what motive we have in that so that's kind of an exciting initiative and now I'm going to stop because I've gone over I apologize so so we uh as you can tell we're very at Georgia State we've been we're in our fourth month right of doing this and I thought that was a very that was very very enlightening Michelle thank you so you have questions for us well it's time I hate this but if if you would like to reach out to any of us and or have a conversation we'd love to talk to you more about staffing and burnout and maybe new roles for librarians and all those things and new roles for our you know what kind of skills would yes boundary negotiations buddy I thank you all very much thank you