 I think it's about time to get started. Welcome everybody. I'm Cliff Lynch. I'm the director of the Coalition for Networked Information. And you're joining us for the last of the project briefings of week two of our fall 2020 CNI virtual member meeting week to just to remind you is themed around the transformation of organizations and professions. I would note two things about week two. We have made some prerecorded videos available as well as the synchronous sessions that you've seen earlier this week. The recordings of the synchronous sessions are going up as fast as we can get them up. And we will be doing a sort of a summing up synthesis Monday at four o'clock which I will lead for week two week three will start thereafter. A couple of logistical things we are recording the session we will make it publicly available after. We have closed captioning available. There is a chat and please feel free to use the chat. You're welcome to introduce yourself in there make comments as the session proceeds. There's also Q&A tool at the bottom of your screen you can use that to print to post questions at any time. Now the way we'll handle the quest. Well, let me back up a moment. We actually have two separate presentations one by two folks from OCLC and one from two folks at Ithaca SNR. We put these together in one session because both of them describe studies that deal with much the same area and we felt that it would be quite valuable to be able to ask questions across the two studies. Many of these studies have both of these studies have gotten quite a bit of attention, and you may have noticed they've been mentioned a number of times during the meeting so far by other presenters. So at the end of the OCLC talk we will take a very brief pause and if there are pending questions that are specifically about that, we will field them. We will move right ahead and we will hear from the Ithaca folks and then we will have Diane Goldenberg Hart from CNI moderate a question and answer session across the two presentations. So with that, both of these presentations are broadly about relationships and partnerships in research support services. These are, this is a major area of concern, a growing one as research becomes increasingly digital in its nature and its products. And I'm not going to say much more about it. We will hear from Rebecca Bryant and Brian Levoie at OCLC research about their study and then subsequently we will hear from Rebecca Springer and from Jane Radecki at Ithaca SNR about the work that they've done. And so I will now turn it over to Brian Levoie to start the presentations. Let me just say a big thank you to all of our presenters for joining us today. Over to you, Brian. Thanks very much, Cliff. And thanks to everyone for joining us today. So in this talk, we're going to give an overview of our new report, social interoperability and research support from OCLC research. In this report, we introduced the concept of social interoperability and we use it to explore cross-campus collaboration in research support services through a series of interviews that we conducted with practitioners representing a wide range of campus stakeholders. And in these interviews, we gathered information about their goals, their interests, their expertise, and of course the importance of cross-campus partnerships in their work. The team consists of myself along with my colleagues, Rebecca Bryant and Annette Dortmund. As Cliff said, Rebecca and I will be presenting to you today. A roadmap for the talk is pretty simple. We'll start by talking a bit about why cross-campus collaboration in research support is important. We'll talk about who the campus stakeholders in research support are. We'll talk about what categories of research support services provide useful illustrations of social interoperability in action. And then we'll finish up with some practical advice that we gathered from our interviews on how to achieve social interoperability in campus partnerships. Let's start with the why. And we begin with the definition. So when we say research support services, what kinds of services are we talking about? Well, this is the definition that we came up with. And as you can see, we're focusing on services having to do with productivity, with analytics, and with discovery and access, which is a pretty broad sweep of services. But to make it a little more concrete, think of things like research data management, research information management, analytics, personal identifier systems, all of these and many more would fall under our definition of research support services. So the starting point for our project is the idea that in many circumstances, cross-campus collaboration is essential for developing and sustaining research support services. And just for some context, take a look at this recent posting for a library chief data strategist position at the University of Rhode Island. And notice the emphasis on the need to work with both internal and external partners and to cultivate relationships in order to make the position successful. This really captures in a nutshell the key premise of our research, which is that research support is an enterprise task. And it's a capacity that's developed by the collaborative efforts of multiple campus stakeholders. And to underline the point, here's a quote from a practitioner we spoke to, where they say pretty emphatically that you really can't get anything done without partnerships. So partnerships and collaboration are important, but the catch is that creating these partnerships and collaborating successfully is hard in a campus environment. And another quote from a practitioner that speaks to this, and I imagine many of you can attest to this as well from your own experiences. Now, in order to make cross-campus partnerships work, you need something we call social interoperability. And this is the second definition I want to bring to your attention. We define social interoperability as the creation and maintenance of working relationships across individuals and organizational units that promote collaboration, communication, and mutual understanding. I'm sure you're familiar with technical interoperability. Social interoperability is like that except here we're talking about people rather than systems. So interoperability is becoming more important as campus partnerships become more prevalent and more necessary. And it seems to us that too often this concept is relegated to the margins of discussions about campus services. We want to pull it into the spotlight. And you probably have too, that people often talk about how, you know, social is harder, or they say that coming to consensus and working in groups is harder than just sitting down and designing a technical solution. But why is it harder? One reason is the campus environment in which social interoperability has to take place. So William Rouse, he's a systems engineering expert, and he's also a former university leader. And he describes universities as complex adaptive systems, which are systems with systems with features like nonlinear dynamic behavior. So this just means people respond or react in disproportionate or ad hoc ways. It's populated by independent agents, people have a lot of freedom to be self directed in their research and in other behaviors. They can even openly oppose organizational endeavors. Their goals and behaviors can differ or conflict disagreements are common. People can operate across purposes. We have intelligent and learning agents. So not only are people independent agents. They're smart independent agents who can develop strategies to thwart one another. You see a lot of self organization, you know, certainly in the university there is hierarchy, you know, within colleges, schools, departments and so on. But there's also a lot of self organization that occurs outside existing hierarchies. And there's no single point of control. Power and decision making is, it's very diffuse across the university and that makes top down mandates difficult to, excuse me, difficult to enforce. So the bottom line is that the campus environment can really make coordinated decision making and collective effort challenging. Or as one person we spoke to put it, it can be like herding flaming cats, which is what they said about their experience as they attempted to implement a campus rim system. So that brings us to our projects beginning with the premise that social interoperability is an important part of successful campus partnerships. We scoped a project involving nearly two dozen interviews with individuals at 17 research intensive us universities and each of those individuals working in a campus unit. That was a stakeholder and research support services. So in these interviews. The individuals discussed in detail what they do. And also their experiences with cross campus partnerships, both in terms of what worked and what didn't work. And from these conversations we conversations we were able to compile a wealth of information and practical advice on achieving social interoperability to support campus partnerships. Okay, so let's turn to the who aspect of social interoperability who are the campus stakeholders and research support services. And answer this question. We developed a model that provides a visual summary of six major campus areas with a strong stake and research support, and therefore the ones that are most likely to be involved in partnerships in research support services. Now by highlighting the parts of campus that stakeholders and research support. This model kind of helped us identify the areas from which to draw our interviews. Now I should point out that these are broadly defined areas they're not necessarily mutually exclusive. The distinctions between them are really ones of focus rather than you know ones of clear administrative boundaries. And obviously this is not a complete model of the campus it's limited to the areas most relevant to research support. But what we tried to do is flesh out each of these areas through our interviews in order to gain a deeper understanding of what people situated in these areas do. And what their interests and challenges are, especially in relation to research support so let's take a few minutes I'm going to run quickly through each of these areas just to make sure that we're all on the same page in terms of understanding what they represent. So the first area is academic affairs. So these are the roles on campus that are responsible for overseeing the teaching learning and research activities at the university so examples of roles in this area would include, of course the provost who's the chief academic officer but then also deans and directors of colleges, schools, institutes, department heads, directors of graduate study and so on. The next slide there is probably our favorite one from the study, and it really nicely illustrates the knowledge gap that can exist about potential collaborative partners across campus. We're happy to report that that comment was actually made by a provost. The second area is research administration, and we've lumped a lot of activities under this heading, but the common thread is that they specialize in advancing the university's research mission through things like helping to identify and secure external funding, developing institutional research strategy and policy, assisting with grant administration, oversight of research conduct, all of these things. And these functions are often collected under a university office of research or something with a similar name, led by a vice president or vice chancellor. Now the quote on the slide from one of our interviews really captures a key goal of research administration units. It's helping researchers focus on conducting and communicating their research by lightning some of the administrative and logistical burdens that are attached to conducting and communicating research. The third area is the library, which is certainly a familiar campus presence with a familiar mission to connect students and faculty with the information that they need for learning and research. But our interviews really highlighted the diversity of library roles and responsibilities supporting the university research mission. Things ranging from the university press to digital humanities institutes to scholarly communication to departmental liaison work. So as these roles suggest today's academic library is deeply embedded in all phases of the research lifecycle. So the quote on this slide indicates there can be a lack of awareness on the part of other campus stakeholders of the library's expanded role supporting research, which is an obstacle that has to be overcome information and communications technology campus units that support a wide array of technology needs including those related to research so for example stores resources high performance computing digital collaboration tools specialized research The interview quotes on this slide highlight one of the key issues that ICT staff tend to face in regard to research support services. Should those services be centralized at the campus level or should they be operated with an individual campus units and that in turn influences who the appropriate campus partners are for ICT staff and what kinds of collaboration are needed. Next faculty affairs and governance. So these are campus units that support faculty members and their careers and scholarly activities and also support faculty participation in university governance. So here we're talking about things like tenure and promotion sabbaticals startup funds faculty searches, and of course the faculty senate. Now the quote on this slide is a good illustration of one of the insights our interviews revealed about individuals working in this category. They often occupy roles that cut across the campus stakeholder network and that makes them especially helpful in making cross campus connections and in convening cross unit venues for discussion and problem solving. And then finally the last piece of the model is communications so communication specialists are responsible for promoting marketing or otherwise raising awareness about university programs accomplishments initiatives. They might be concentrated in a university communications or public affairs office, or they could be distributed across campus and different units like academic units corporate relations offices the research office alumni relations and so on. Now the quote on this slide from an interview suggests that campus specialists are particularly skilled in building networks and building community. And that makes them valuable partners in almost any kind of cross campus collaborative activity. So that's a quick overview. There's a far more detail on each of these areas in the reports I'll refer you to that. But the big takeaway here is that the first step in achieving social interoperability is to know your partners. What do they do. What do they prioritize how do they see themselves contributing to the university mission. So that means looking beyond traditional or superficial perceptions of what these campus units do in order to really understand how their responsibilities evolve, how they expand and how they might be reprioritized over time. So we've covered the why and the who and I'm now going to turn it over to my colleague Rebecca to take us through the what and how of our project findings so Rebecca. All right, thanks Brian so we're going to talk a little bit about the what is Brian forward so to the next slide. And within the report, and not only do we talk about who those stakeholders are we also talk about them in context in these four areas Brian mentioned earlier research data management, research information management, research analytics in a door, research data adoption. And next slide please Brian, we talked about really examining what the stakeholder interest is, and at a quick glance what you can see here is that there's a lot of stakeholders who all of these stakeholders have an interest in, at least a couple of and many of them have an interest in all of them. And, and were we to examine other areas of research report, research support we would have found more. And, and next slide please Brian. There's, we also go into a lot of depth about these four areas. In the interest of time I'm not going to do that in that presentation in the presentation today you can. I encourage you to go to the report and dig into the different sections. But within the area of research data management. We found the pressing need for cross institutional social interoperability and a lot of investment for across campus. And this could include things like having a data librarian embedded in the research office, or perhaps having a research data service that's housed housed in the library, but funded by the Office of Research so, so there's there's role sharing there as well. There's lots of partnerships between the library and research development officers who are a sub component of usually the research office, providing training helping with DMP tools, and etc. And we heard this quote from one of our informants who said, I don't think that either our research data management services, or the campus room system would have been as successful as library only. It's critical that they were backed by the Office of Research, because it's helped keep it to be more of a campus wide perspective. Next slide Brian. One of the things that we want to share with the community, and especially to our OCLC library community, or comments we heard from outside of the library from the stakeholders we talked with about how they perceive the library. We heard many positive things. For instance, we heard that the, the library's value is a trusted and agnostic part partner for projects and also add for sustainable projects, it's a place for things to go and to be stewarded. There's also expertise that's recognized in the form of metadata, metadata management, metadata cleanup, licensing, working with vendors for many of these systems like research data manager repositories or RIM systems. And also there's a lot of expertise in libraries related to metadata and then using that for research impact and bibliometrics. Next slide. But we also heard the other side of that and we want to bring this to the, to our library community is some things that some of our stakeholders said, I, maybe, maybe this is something that we can all work together on. And some criticisms that we heard, for instance, wasn't that that libraries can sometimes over emphasize values, which means that they may not be communicating as effectively to what the other unit, the other unit cares about. And that there furthermore, the value proposition of what the library has to add may be diluted because it's very interested in being a service unit, but also can risk be provide sort of being trying to be everything for everybody. We are concerned that the library is slow moving. We also heard a fair amount about imposter syndrome and a lack of confidence among librarians that our informants also really felt was unwarranted. They value library skills and they want librarians to come forward as full partners. They also signal to us a discomfort with finances is that often, you know, things cost money and they often thought that that libraries could talk about an unrealistic desire for everything to be free. Now, the two quotes I have here on the right actually didn't come from our study. They came from a more recent series of discussions that OCLC has hosted with the European Association of Research Libraries, Lee Bear. And they also brought up that, yeah, we're pretty mission focused here. But if we were a little more neutral, and we tried to be a little bit more helpful, and we tailored our message a little bit better, we might be more successful. So next slide Brian. And we're going to talk about the how next slide. So, and I'm going to have you go ahead to the next slide Brian. We have here finally a section of our report that is really about how do we do this. This doesn't seem like rocket science. A lot of this you probably already know a lot of this I already knew. But yet, even in my work here at OCLC and my colleague and co author and at the door minutes of this too. It's useful to go back to this list to remember. So we go, this is advice that's really gleaned from our informants, and we encourage you to, you know, think about how our language may and our use of jargon in one part of campus may not collide with another. And as Nina Exner and Stephen Bollinger said in their presentation yesterday, it's also really important to offer concrete solutions to problems. So I'm not going to go into a lot of detail about here but there's a lot of rich information in our report. So I'm almost at the conclusion and Brian's going to forward here to the next slide is that in addition to the report. We've been continuing to try to learn together with the OCLC research library partnership this fall. So one of the things that we've been offering is a webinar series for our RLP members that began in August and will come in go on through February. We've called upon the expertise of our RLP members such as the University of Miami Syracuse University and the University of Cincinnati and Arizona and Illinois furthermore to talk about the way that they may be collaborating with other parts of campus through shared positions embedded libraries and interdisciplinary research teams and bibliometric and research impact. There are a lot of stories. So in addition to the report, if you want to dig in more about one of these areas, the webinars themselves are available to our RLP partners live, but the recordings are available at the link that's here. And so with that, I believe we conclude. And we encourage any questions that you might have. Thank you. Thanks, Rebecca and Brian. I'm not seeing any questions in the Q&A at this point. So please feel free to pass the baton on to the team from Ithaca SNR. Great. Thank you guys. So hi everyone. Thanks again for joining us on this Friday afternoon. I'm Jane Radike and with me as my colleague Rebecca Springer and we're both from Ithaca Esplosar. Our presentation today has a similar topic to what Brian and Rebecca just discussed. Today we will be discussing the provision and organization of research data services within US higher education. Next slide please, Rebecca. This presentation introduces a report published earlier this week that quantifies different types of research data services offered across libraries, IT and research computing departments, professional schools and more on 120 college and university campuses. We will be sharing about the problem we were trying to solve in creating this report, the methodology we created and used, and some of the high level results and some key conclusions and takeaways, and we invite all of you to read the report in full at the link provided. Next slide please. Before we go any further and dive into the methodology and results, it's important to explain what we mean by research data services. So for the purpose of our study, we defined research data services as any concrete programmatic offering intended to support researchers in working with data. The goals of this are consultations in training, finding, managing, analyzing and visualizing or sharing data, including how to use specific software or programming languages to work with data. We have more details on what we included and excluded as services in the report. Other words, the definition of research data services that we worked with focused on support that directly helps scholars advance the research agendas. Next slide please. In the beginning phases of this research, we noticed that there is extensive literature documenting support for research data management and other services provided by the library and institutions. However, to our knowledge there has been no systematic quantitative study of research data services provided by US academic institutions holistically. The vision of research data services varies significantly from campus to campus, and we need a more complete picture of research data services in US higher education, in order to begin talking about the best ways to deliver, coordinate and improve services. This is where our design methodology played a key role in our research. Thanks Jane. I'm going to talk a little bit about the methodology that we used, both to give you some context for the results that Jane is about to share with you, but also because we hope that by sharing about our methodology we can be of use to the community more broadly and maybe some others might find it useful in their own work. We have a lot of details about the methodology in the report itself. But just to give you an overview of it. When we started looking at the problem of quantifying research data services, we realized that traditional methodologies like surveys wouldn't be a good fit for this particular problem because of the highly decentralized organization of these services across institutions. There's generally no one purpose person on a college or university campus who is able to speak accurately to all of the different research data services that are available across that campus. So for that reason, we devised a web based inventory methodology that we use for this project. This was essentially a systematic way to look through and scan institutional web pages for evidence of research data services, and then input that data in a structured way we used to Google format of convenience. So we developed this methodology through an initial proof of concept test with about 30 institutions refined it and then Jane actually implemented the entire thing did the whole inventory for 120 different institutions. We looked at 40 each in the three Carnegie categories here are one, which are your sort of most research intensive universities are twos and then what we're calling slacks or small liberal or selective liberal arts colleges. We did look into looking at other classifications as well but in the proof of concept test it was looking like there wasn't going to be a sufficient number of research data services at those institutions to warrant further study. And it's probably worth saying that this was an extremely labor intensive process. It took between three to five hours per our one institution and but around two hours per our to or slack institution in order to complete the inventory. Really briefly we classified the services that we found, primarily based on where we found them within the organizational structure of the college or university. So, we found services in the library, a category that we're calling it department but it does include research computing units as well. Independent research centers and facilities that includes core facilities, academic departments, the medical school to business school and other professional schools. Within the library and IT department categories we further classified services by what type of service they were as well as any subject focus that they had. And then within the other organizational locations we classified services in the seven categories that you see in the list to the right there. And we have detailed definitions of all of these categories in the report itself. Based on this it is important, I think to air upfront what our data can and can't tell us before Jane dives into the results. So obviously in this research we're using web presence as a proxy for institutional legitimacy. We think this is a reasonable approximation to make but of course it is not airtight. So we can't find through this process services that aren't advertised online for whatever reason. We also recognize that a lot of research data services are provided by people in roles where they might have just, you know, a research data services as a small part of their role is not advertised online faculty may be getting support from peers collaborators postdocs external consultants. These are all really important and how researchers are working with data but they're not the focus of this research and not captured by this methodology. We obviously don't know how how much a service is used or how effective it is only that it exists. We don't claim to have 100% accuracy in our inventory of research data services on an institution by institution basis we don't have detailed local knowledge we only have what we can figure out from looking at web pages. We do think that that that the aggregated picture is on the whole pretty representative. Finally, you know, to sum this up, we designed this process with the aim of identifying the largest number of research data services as comprehensively and efficiently as possible. We think we came up with a solution that's kind of the best that you could do for this particular research problem. But we recognize that it's not, it's not perfect and that there are ways that the methodology could be improved so we welcome engagement on that. And we have made our full data set available and welcome you to take a look at it. And with that I'll turn things over to Jane to share some of the results. Great. Thanks Rebecca. So with that methodology that Rebecca just described in mind we can start to dive into some of the high level results that we pulled out. Next slide please. So this is a graph of the overall number of services we identified being offered by Carnegie classification. As you can see it shows that the distribution of research data services across even the relatively well resourced institutions, sampled in this inventory is extremely uneven. There is a total of 478 unique research data services that we were able to identify across our sample of 40 R1s, 40 R2s, and 40 slacks. And unsurprisingly, the majority of these services were found within R1s. In fact, R1 institutions exceed R2s and slacks by more than double in the number of research data services offered. This leads to R1s having 63% of the total services found or an average of 7.6 services per institution. R2s having 22% of the services or an average of 2.6 services per institution and slacks having 15% of the total services or an average of 1.8 services per institution. Breaking these results down by organizational location or what part of the college or university is providing the service helps us to shed some light on the gap between R1s and other institutions. Next slide please. For some types of data science and research high performance computing is either essential or highly desirable. While the way that we ended up defining research data services for this research project did not end up including the provision of high performance computing and our total number of services. We did actually end up recording whether each institution we sampled had high performance computing and where this service was located within their university structure. As you can see from what we could identify shown in this graph here, it shows a significant inequity in campus mediated access to high performance computing across Carnegie classifications. 100% of sampled R1 institutions offered high performance computing facilities, as did 60% of R2s and 28% of slacks. This finding shows that researchers within R2 and slack institutions may possibly be limited in the breadth of resources when it comes to resource intensive computing, and it may limit what they can undertake. This may also lead them to turn to third party providers as well. Next slide please. As central as campus units, libraries and IT departments are well positioned to offer research data services to scholars from a really wide variety of fields. And we found that libraries are important providers of research data services across all Carnegie classifications. Libraries in fact are the largest contributors to research data services on R1 campuses, offering 32% of the research data services we identified at these R1 institutions. The average R1 library offers 2.4 unique research data services. By contrast, the average number of library research data services offered is less than one for both R2s and slacks. The most commonly offered service is generalist consultation. A small number of R1 and slack libraries offer research data services with a variety of disciplinary methodological and technical focuses. We also found that the profile of the types of research data services offered within libraries was roughly consistent across Carnegie classifications. However, as you can see from this graph here, there are 20% of R1s, 68% of R2s, and 60% of slacks that do not offer any research data services in their library. Next slide. So, although IT departments and as Rebecca mentioned earlier, this also includes research computing units provide fewer research data services than libraries overall. They are still an extremely important service provider, particularly on campuses where the library does not offer research data services. While the library leads in the provision of research data services, some IT departments also offer training in statistical software, research data management consultations, and other services. In some cases it does appear that services offered by the IT department may actually be filling the gaps of what the library would typically provide. For example, 87% of R1s that do not provide research data services through the library do through their IT department. Next slide please. So, a wide variety of research data services, most focused on specific disciplines or methodologies can be found in academic departments, independent research centers and facilities and professional schools. Most services offered outside the library and IT department have a specialized disciplinary or methodological focus. These services most commonly focus on statistics and at R1 universities, bioinformatics, and the average R1 has at least one of each of these services. Statistics services are also important at R2 institutions. Statistics, as you will notice from the graph, was the most common type of research data service offered outside the library or IT department. And these services focused on providing support for statistical analysis and methodological design. The prevalence of these statistical services across the university structure suggests that there is a strong demand for statistical support across multiple disciplines and user groups. Next slide please. So, while statistics and bioinformatics research data services are common at R1 universities, the picture for other Carnegie classifications and other types of services is very different as you can see from this graph. More than half of our two institutions offer at least one statistics focused service around half of our ones and one third of our twos and slacks provide geospatial research data services. Services focused on clinical data, social sciences, business or the humanities are each found at less than 20% of institutions across Carnegie classifications. With the research data services found outside of the library and IT department, 37% of these services were actually statistics and 26% of them were bioinformatics. So, from our analysis, this shows that there may be severely disciplinary and methodological areas outside of the library and IT department which may be under provisioned with research data services across institution types. Next slide. Thanks. I'll just provide a couple of closing thoughts. We wanted to present our takeaways from this research as questions to you all. We view this research as a starting point for further investigation and we're interested in in thoughts or experiences that folks in the audience have that that may speak to some of these questions. So some of the things that we're thinking about, based on our research, of which there's a lot more in the report Jane shared a little sample of what we found include, you know, thinking about the currently very decentralized model of how research data services are provided on many campuses and what the advantages and disadvantages of that are. There may be advantages in terms of faculty having local access to research data services that are sort of key to their disciplinary or departmental cultures, but there may be inefficiencies associated with that so where's the balance there. Secondly, we know that you know many libraries and IT departments are serving as sort of central hubs to connect faculty with different types of research data services that they may need and I think this was spoken to a little bit in the OCLC presentation. To what extent is that the right role for libraries and IT departments should they be focusing on developing their own research data services, or is there a way to marry both of those functions. Thirdly, obviously we have seen great inequities in the extent of provision of research data services between different institutions. Given limited resources we're interested in the potential for collaborative models across different institutions to allow greater service provision at an efficient scale. The data curation network is one example of an organization that's trying to do this and is being successful at that so far. Finally, we do have of course a lot of questions about how many researchers what types of researchers use different data services how this compares to demand and what the effect of those services is on their research. So this is a topic for further research. In many different ways, we do want to highlight that we have another project ongoing at SNR focused on understanding support for big data research and how that can be improved. That's a project in collaboration with 21 academic libraries and it'll wrap up at the end of next year and hopefully provide some insights toward answering some of these questions but there's a lot more work to be done regardless. So with that, I will have one slide about further reading. We are doing a lot of work on research data and the research enterprise more broadly at SNR, some of which has been being presented at CNI. So feel free to go back to our slides at your convenience and take a look at these links if you haven't seen this work already. And with that I will hand things over to I think either Diane or Cliff to moderate the Q&A session. Thank you. Thank you for that to everyone. Very stimulating conversation and very complimentary projects here so it was very interesting to hear both of you present one after the other and think about them in tandem like that. And I will just say to echo Rebecca's slide that there is a wealth of information from both OCLC and Ithaca SNR on this topic and I urge everyone to go check that out. It's really good stuff. So the floor is open for questions. Please type your questions into the Q&A box. And we have a question to start out now from Chuck Ekman and this is for all of you. Do either of the studies offer insight into the factors that influence institutional choices or patterns regarding the placement of certain research services either in the library or IT, ICT. For example, RDM, statistical software support, RIMs, infrastructure, etc. Is it random? That's definitely a question that we have based on our research and, you know, doing a snapshot of what's available on the web at one given time. And we can't tell us that. I think Jane can probably attest just how random it is and how hard it was to find patterns or to predict where certain things would end up across organizational structures. So it seems to me to be an area for further research. I will jump in and definitely echo Rebecca. It's definitely a question that we came away with when I was going through and kind of doing the inventory and trying to find all the services. You'd kind of be amazed at how many services were sparsed across the kind of institutional structure in really odd places. Especially like some professional schools like nursing schools or dentistry schools, you find them in really odd places. You would think that you kind of would get the pattern down and for most libraries that they seem to kind of all generally kind of be in this the same area ish, I would say, but for the remainder of the services that were outside of the library. And that includes the IT department is like, I'm including the IT department in those specialized services as well. The placement of them was sometimes really bizarre and quite random. But so definitely definitely a really good question to raise and one that we came away with as well. This is Brian from OCLC. Yeah, this is something that came up in our our study and actually came up in another study we did called the realities of research data management where we were looking at research data management services at a number of research intensive universities in different contexts. There's a couple points about this. You know, the question is says about institutional choices and you say choices. It's it's the there's some implication of intentionality. And oftentimes that intentionality really is lacking and it's other factors that drive these decisions. So for example, we've seen that some of it is personality driven if you have a strong leader in a particular campus unit who is a visionary and you know believes in the importance of research data management services. You know they they they might drive the creation of those services within their within their unit and we saw that with a number of research data management services at different universities where it was a there was a strong leader who initiated their implementation and that's why it ended up there. Another issue that came up in this we heard in our stakeholder study from a number of interviewees. There's a territorial issue involved too. Sometimes, you know, some of these services sort of arise in sort of an ad hoc way. There's a research cohort that uses a particular service that specialized to their discipline. And there may be opportunities to sort of raise that service up and centralize it and make it available to others. But once that service is established in a particular unit there there's reluctance to give it up if if people from Central ICT come in and try to, you know, engage with these local units in terms of oh can we centralize this service it's sometimes seen as an invasion in a way of of the territory of that unit and they want to you know they say you know you don't understand us we have special requirements that you can't meet the service has to stay here. And it was really interesting to see in talking to some of our our campus computing interview interviewees that you know they you know they said look at our campus, all these different academic units they have their own it staff that manage their own systems and and they're not going to give that up even if it makes sense to raise it up to a to a centralized level so so the point being that you know and exploring this question I think it's a really good one. You know, I think we see a lack of strategy in making these decisions and that the the patterns of central versus local that we see are really sort of a ad hoc random process to a large extent Rebecca I think you wanted to add something about rim systems. Yeah, this is also something I just wanted to sort of give a shout out some for some work that we're getting we're, we're just launching related to research information management systems we're doing some case studies related to rim practices. And I think that most of you who are attending if you have experience with this know that this could be in the library this could be in the research office this probably is also in a number of colleges. And so part of what we'd also like to know is that's a question you're asking is why and stay tuned we'll publish on this next year. Thank you. Thanks. Thanks Chuck for that question. And to all of you for your thoughtful responses I, it makes me think of those projects that sometimes pop up because there's a faculty project that might drive the seed that then leads to the implementation of service. Can I also add that I actually think that this is where in, you know, Brian's part of the presentation he talked about this model from Rouse, that I really, for me this was like the light bulb moment of finding finding his description of higher ed and why it's, you know, because I spent before I joined OCLC I spent many years as a university administrator in academic affairs and it's like why is it so hard and it's, I found that description it's like, oh, that's why it's so hard. And so I really encourage you that many others if you haven't seen this before I didn't read this, you may have the same sort of light bulb moment epiphany sense of grace almost that comes over you for understanding why it's so hard. And that I think also then can can help us know how to work within that but it because it is so decentralized and everybody's off doing their own thing it's filled with smart people with their own agendas. And that's what happens. It's, yeah. Thanks Rebecca. I think Cliff has a question now. Also, thank you. So this is just a quick one and I think it's mostly for for the Ithaca folks. I found that your way of fascinating things, fascinated things into discipline specialized and then methodology specialized very useful. But what I'm wondering is, I didn't see amongst the methodologies data science, yet I've been following, you know, efforts in many, many institutions to figure out how to do data science as a sort of a supporting methodology is distinct from the path data science as a academic major or graduate degree or something like that. Is your sense that that is still in the in the classification you did is it just kind of subsumed under statistics as an extension of statistics, or was it just not visible enough to merit its own category or what. That's a good question. I think so there are a couple of things at play. One is what you're getting at that we have a lot of, you know, center for data science, or equivalents which are, you know, providing, in some cases, support to members of that academic research center. We did not include in our analysis because it wasn't that support was not available more broadly to the campus as a whole so if a research data service was scoped only for the members of a particular research center, sort of as an incentive to be a part of that research center for instance and that wasn't included. Another consideration is as you mentioned statistics is kind of data science now and so those blend together. And we found, you know, we generally that what was left after you sort of take out those two categories was was often focused on one particular disciplinary or methodological area. So, you know, focused on social sciences are focused on visualization or focused on bioinformatics outside of, of, say the library where you see some more generalist, you know, services and in terms of, you know, working with data one on one. I see a lot of, well, it's hard to find something that is data science but not statistics and not specific to a particular discipline. The Venn diagram just doesn't leave very much in the middle. So I hope that that makes sense as a explanation. I don't know if Jane wants to add anything. That's very helpful. Thank you. Yeah, I'll just I'll add on to what Rebecca was saying. The Venn diagram that she's talking about it really doesn't leave much space for any specific services, but as Rebecca also mentioned, a lot of data science today involves statistics and essentially the process that we developed in our methodology was to first look at the library and then at the IT department and to Trev's question of is there any method of where services where I am, I always went to look for data science programs after the library and the IT department because I knew that that was one area that would most likely have services and that pretty much held true for most of our ones. So that was one area that I kind of knew and could count on would have services. The only caveat for our research as as Rebecca mentioned, was that if, for example, statistics data analysis services for for data science was only offered to say students or grad students in data science program, we didn't end up counting that service, we would note it like we noted it down in a note section. But we were only counting services that were offered to kind of the entire broader university structure. So, there definitely are data science services that are kind of lumped into that statistic services because a lot of them are coming from those data science areas and disciplines and it was kind of pretty interesting to see that that was the one area that I could count on the most for having some services, the other area that I could count on the most for having services was digging through medical schools and finding bioinformatics services through through medical schools. That's really interesting. Thanks. Thanks for the for answering those Jane and Rebecca and thanks Cliff for that question. And I see now that we are a little bit past time, and it is Friday afternoon here on the East Coast anyway so I don't want to hold everybody up any longer, but with sincere thanks to all of our presenters for bringing your talks together in this single hour we really appreciate it. And all of our attendees thank you so much for spending some time with us here at CNI. I'm going to go ahead and turn off the recording now, but any attendees who are still with us who would like to join us have a little chat, maybe ask questions of our presenters or make some comments. Please feel free to stick around just raise your hand and I can turn on your microphone. And with that I wish everyone a great weekend. Take care, and we'll see you next week. Bye bye.