Digital Frontiers 2014: Developing Library Services for Digital Humanities & E-Science Support Using Qualitative Research

Editor's Notes

• #2: RAFIA One year ago, the UT Arlington Library underwent a total staff reorganization, during which a department of scholarly communications was created, consisting of a Director, a Digital Humanities Librarian, and a Data & E-Science Librarian. These roles were new to the library, and were positioned to provide services to the university that had not yet been offered in a formal, centralized way. Our first step was to understand what kind of support faculty needed in the areas of digital research support and data management and curation. We conducted a series of interviews across campus, including the Vice Presidents of the Office of Information Technology and Research Administration, deans of colleges, and several faculty members. We asked questions on what technologies researchers are using, what technology support they need, where they store their data during and after research projects, their experience with data management plans, the role of digital projects in the tenure evaluation process, and attitudes on data sharing and open access. We then transcribed these interviews, identified patterns and common themes, and classified each as a strength, weakness, opportunity, or threat (SWOT). Our poster will cover the entire research process including recruitment and interviews, highlight the themes and findings, and present how we used the analysis to develop our services. We underwent another smaller reorganization at the beginning of this semester wherein the scholarly communication department was absorbed into the Outreach and Scholarship Department.
• #3: CLARKE
• #4: CLARKE
• #5: CLARKE
• #6: CLARKE
• #7: RAIFIA
• #8: RAFIA I think of you as the people who are the navigators. You know where the information is, you curate and manage the information, you can help people find what they need, you can show them how to use it in the most effective ways. And you need the technology tools underlying that to be able to enable that role. The library is the heart and soul of the university”
• #9: RAFIA “God bless the social science librarian. So I’m hoping that the specializations, and the special help will always be there, particularly for students—the budding researchers. They’ve just got to have that help at the front end. And I know when I was teaching I always had the librarian that was in planning, dedicated to the social, applied sciences, come in and talk to the brand new masters students and work with them for a couple classes, and it just made a world of difference. So I feel strongly that the specialization should be there, that that kind of help should be there, and probably almost mandatory that there be these interactions, and give librarians a bigger part at the front end of a semester with students...Yes, I would hope all kinds of things like that will continue because that demand will always be there. People don’t come to college with these skills in place.”
• #10: RAFIA
• #11: CLARKE
• #13: CLARKE “We have a very inadequate research environment right now… right now, just from a budgetary standpoint, [supporting research is] about 1/36th of OIT’s budget. And that’s us going out and assigning a percentage of the infrastructure costs to research. Direct research expenditures I have $200,000, out of an $18 million budget. So I am not funded to support research yet.” “I also believe that we have pretty far down the totem pole computers and computer support… we, as a university—have not put the money into it that needs to go into that sort of thing” “From my standpoint, the first core is the infrastructure that everyone depends on. That begins with the network… The lights have to stay green. And, it’s also wildly underfunded” Faster Bandwidth/transfer speeds “Our main bottleneck that we are facing is the speed by which we can transfer the data” “For the scale of the data that we have, bandwidth is a huge problem. As an example, many of us are having our sequencing done at Southwestern. It’s way faster to drive over there with a hard drive and drive back than it is to move it up by the internet”
• #14: CLARKE “We have a very inadequate research environment right now… right now, just from a budgetary standpoint, [supporting research is] about 1/36th of OIT’s budget. And that’s us going out and assigning a percentage of the infrastructure costs to research. Direct research expenditures I have $200,000, out of an $18 million budget. So I am not funded to support research yet.” “I also believe that we have pretty far down the totem pole computers and computer support… we, as a university—have not put the money into it that needs to go into that sort of thing” “From my standpoint, the first core is the infrastructure that everyone depends on. That begins with the network… The lights have to stay green. And, it’s also wildly underfunded” Software Support “for me, the data analysis needs have not been quite what I had hoped…… I cannot impress upon everyone enough the need for those two toolpacks in SPSS because without them, it’s kind of like giving people half of the package, but not the full package. I think we need to be looking into more software for faculty across the university.” Faster Bandwidth/transfer speeds “Our main bottleneck that we are facing is the speed by which we can transfer the data” “For the scale of the data that we have, bandwidth is a huge problem. As an example, many of us are having our sequencing done at Southwestern. It’s way faster to drive over there with a hard drive and drive back than it is to move it up by the internet” Storage Support “The Department of Biology doesn’t have its own server that’s capable of doing genome analysis, which is a third of this department. It is left to the individual labs to secure that computational infrastructure.”
• #16: CLARKE
• #17: “I think that a lot of what we do actually is based on primary data that’s gotten from the federal government or from the state government, and even some local data that probably has to come from the cities themselves; I’m not sure if that comes out of a library. But certainly primary state and federal data would be all essential”
• #18: RAFIA
• #19: RAFIA
• #20: RAFIA “I’m also working on compiling a corpus, and that’s sort of another set of issues involving access to materials, and licensing of materials, and reuse, and what’s copyrighted and what’s newly created and spoken, and so on…what percentage of materials that will be made of really greatly depends on this copyright issue” Communication (cont.) “I would say open access is kind of like the data management thing in the sense of: the landscape is changing, faculty are not necessarily as aware of everything that they can and can’t do; what the publishers will sign off on, and what they won’t. Again, any help that you can give in terms of advice, and “This is how you get it done,” kind of thing.”
• #21: CLARKE
• #22: CLARKE Researchers perceive a need for the archiving of research data for a number of reasons aside from mandates “what I could envision is if there was a regulation that for each publication that I submit I will have to deposit all data, including methods and so on, into a folder and give this to the library. This is something I would find useful in part also to us as researchers.” “The American Economic Review, where I was an associate editor, has a policy which says, when we publish a paper, you provide us the data. Now, again, there’s exceptions: if the data’s owned by someone else, you don’t provide it. You only provide the data that’s reported in the paper, so if you have other data, you don’t provide that. And, you know, exactly what format, and what time—all these things need to get worked out, right?” “I know some journals are starting to—right now I think just ask optionally if you want to include the dataset so that other people can compare and access it, but I’m pretty sure the move is going to be towards making that more and more of a requirement. But I also know that, from the journal’s point of view, part of the problem then, as well: does the journal have the server space to keep all of that, and to make it eternally accessible to people with different formats?” “the fact that federal funding organizations are now requiring this, I think, is pushing universities to develop storage solutions, and sharing norms, and a template that you can submit for your data management plan, which is probably a good thing” Benefits of data archiving (cont.) Concerns about losing data & having someone responsible for backing up & preserving “That’s a huge, huge unsolved issue for us even within our own field, data archival, we are constantly collecting data and we run the risk that the older data gets neglected, forgotten and lost to the point that where we are so focused on the new data coming out from the experiment that’s where the excitement is, that excitement of discovery, that we lose our ability to even go back to the data that we collected five years ago in 2009 and that we have no way to reinterpret it, to reanalyze it, to relook at it.” Befits of data archiving (cont.) Making data accessible that would otherwise be lost or not kept “if I sequence a new genome—I’m uploading the raw sequence reads, which is the baseline data. So, somebody could recapitulate the analysis if they had a detailed enough description of the method and also an assembly, that’s like the final product. But all the stuff in the middle is potentially lost. Like I said, for that, maybe 20% is the raw data and somewhere between 5 and 10% is the final product data. The other 80% is the analysis steps, and that’s the stuff that gets lost if there’s no long-term solution for storing all that information.” “if you’re getting federal funding, you’re supposed to have open access for your data. So there are people who are being supported with NSF grants, so they’re getting that push.” “Yeah, I can see help from the library in developing and formulating a good data management plan for high quality proposal, so far I haven’t had to face it, but I do see a need for it, proactively getting ready for it is a good idea, some day and to have some in house expertise so I can pick up the phone and say “I have a proposal that is due in three days, do you have someone in house who can help me with it.” I think that would be tremendous.”
• #23: CLARKE “I had not even thought about [making my data accessible in a repository], and none of my colleagues—I don’t think we’ve even really thought about that. But, for one thing, going up for tenure, it would just be something you can add: a link that shows all of the datasets that you’ve created across the course—because analyzing secondary data is not the same thing as collecting primary data… I really just think in terms of creating a portfolio for your tenure package. You can go there and say, ‘Here’s my dataset depository.’” “the younger generation, I have no clue—they would probably jump at a chance to store [their data] in some more accessible place that others could—particularly if they had to be cited every time someone used some of that. “ Benefits of data archiving (cont.) Sharing data is important for furthering research “I think [making research data available is] important. It’s especially important in genomics. Not only the data, there’s increasingly a push to make your method, because so much of the method is computational, and there are lots of filtering steps, to make those explicit. There is still an awful lot of genomics and bioinformatics research that is not reproducible.”
• #24: CLARKE “I think in our department, we’re pretty comfortable with a range of e-projects…And we’ve talked a lot about how these should count as the kinds of things that go on—not just printed books, but the creation of materials, the archiving, and plans dealing with that are a big part of what goes on…But I think it should count, yes, definitely. I mean, if it’s good scholarship, it kind of doesn’t matter what mode it’s in, right?” “the people who are engaging in [digital humanities projects] want to be certain that full scholarly credit is going to be given to them” . In our department, that certainly counts as a thing you do that contributes to scholarship, but I can’t speak for even the other 11 College of Liberal Arts schools; I think everyone sort of has to figure out what their scholars are doing.
• #25: CLARKE “But I think that as data management plans become required by federal research sponsors, probably a more standardized framework is going to evolve. And so I think it’s going to be important for you to be on top of, what are the agencies considering acceptable? Because that’s going to drive what the investigators want, and going to drive what they’ll ask for from the library.”
• #26: CLARKE “I wasn’t aware that we had [an institutional repository], but I would be definitely in support of [putting my work in there], because definitely it would increase my scholarship’s impact factor, which is one of the things—in five years, when I go up for tenure, that’s one of the things I have to report on is the impact of my scholarship, and one of the ways I do that is, how many times was it downloaded”
• #27: CLARKE “It’s especially handy for undergraduates and things that are accessing from home. For example, I have honors college students in some of my courses, and they are required to write reports, and I show them how to find papers, but sometimes they can’t figure out how to get them from home. You know, if you’re not on the VPN, you don’t have subscription. So that’s a hurdle. If there were links on Google Scholar to get it from the university, then that would reduce the hurdle for some undergrads, I’m sure.” “I think once—of course, the journals have fought this huge battle and finally lost, that the paper is no longer theirs—I mean, you see that with the books and with the journals that they’re finally going, ‘OK, we’ve got to join the 21st century and do all this,’
• #28: CLARKE Factors restricting data sharing/deposit in data repository Ensuring credit will be given “I think, honestly, the constraints around data reuse are not from the publishers side; it’s from the researchers side, who says, “I spent all this time and energy collecting this data and I’m not going to just give it to someone else for them to publish.” And so I think that problem needs to get fixed first.” Factors restricting data sharing/deposit in data repository Fear of getting scooped “About ten years ago I had some unpleasant experiences because I was too openly sharing preliminary results on my webpage; so it was copied and published quickly after that. Since then I am very reluctant to show or to showcase immediate non-published results… on stuff where you can basically get the ones and the zeros—so all the logical bits—I am meanwhile restricted making them publicly available” Factors restricting data sharing/deposit in data repository When to share: Ensuring you get all your research out of data before sharing it “In other fields, even in empirical economics, but in finance, for example, you might spend three or four years of your life collecting this unique dataset. You are not expected to—and it’s not appropriate—for you to share it until you’ve published five papers. That’s your tenure; you get your tenure, you publish five papers, and then maybe, in theory, you’ll share it.” Factors restricting data sharing/deposit in data repository Concerns surrounding data misuse/dealing with sensitive data “Much of our research is human subjects based; especially the experimental part, but lots of the observational parts. And so there are quite strict human subjects constraints and concerns. So the anonymizing of the data; what’s retained; what’s not retained: those really fall under that IRB constraint.”
• #29: CLARKE Factors restricting making the publication open access Journals don’t allow it “many of the professional journals are slow to pick up on that sort of thing, and don’t really allow that kind of post-review” Factors restricting making the publication open access Information access is not an issue so OA doesn’t solve a pressing problem for them “It’s so rare to run into something that the library doesn’t have access to through the journals website nowadays, that I almost never run into a paper, unless it’s really old or an obscure journal, that I can’t get. And, even then, JSTOR has it if it’s old, most of the time. If not, the first step for me is usually email the author, and see if they’ll do it. Lots of people have them linked from their website. Even before I email, I’ll go and see if they’ve got the PDF posted on their website. So, it’s pretty rare for me to run into something that I can’t get.” Factors restricting making the publication open access Peer review concerns “The hard part is how to support peer review when you have OA and I think that the real challenge and the whole idea of peer review is crucial to scientific knowledge. I know in my field I know for that even for anything that I access through OA, I also want it to be honest I have a filter and I won’t even waste my time opening an article unless I know it’s provenance that it came from a particular experiment, or it came from a particular lab, or it came from a particular journal, otherwise, I could just be wasting my time.” Factors restricting making the publication open access Journal publication is a matter of prestige “what we get in return for not getting paid money is essentially getting paid in kind by getting our work into a publisher’s journal.” Factors restricting making the publication open access Economic concerns/publisher’s rights concerns “it is clearly an intellectual property taking. You are taking something that currently belongs to Elsevier and you are forcibly taking it from them. And that’s not legal... What is the intellectual value added by the publishers? Non-zero, for sure, right? They do the review—they handle the reviewing process; they do the layout; I think in some disciplines, like biology, it’s significantly—there’s a lot of editing that gets done, in others there’s minimal editing. But there’s clearly some intellectual input that they put in to these papers. And, at the moment, they own the copyright. And so I think you can’t simply just go and post it without getting buy-in.”
• #30: RAFIA So where do we go form here, what did we do with info? Mini-reorg
• #31: RAFIA So position moved around (like legos) to re-orient, restructure, components (of schol comm) same