UC-DLF 2018: Defining and Sustaining Digital Collection and Scholarship Services
Panel Description:
Researchers across UC campuses are engaging in Digital Scholarship (DS), including building digital collections, with increasing frequency and across disciplines. Such scholarship includes an eclectic set of methodologies [1]; and researchers often rely on the library for needed subject, technical, and preservation expertise. The business models and practices that define library DS engagement are, however, in their infancy, extremely varied, and uneven.
This panel will discuss needed support for DS across a range of activities in order to better understand the challenges we are facing from service level, organizational, and business model perspectives. In advance of the conference, each panelist will prepare a brief, written statement that address the following questions:
- What current use cases demonstrate a need for DS services?
- What are the demographics of the patrons we are serving?
- Who are we missing?
- How are digital outputs changing our collection and preservation strategies and what changes should we make in the future?
- What additional or re-deployed resources and labor will be required to provide necessary services?
- Are current and imagined services sustainable compared to more traditional library services?
During the panel, panelists will participate in a moderated discussion of the major themes that emerge from the written statements. Panel attendees will be invited to join the discussion for the final 15 minutes of the panel. Panelist responses to discussion questions will be published below on February 20, 2018.
Panel participants:
- Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA
- Quinn Dombrowski Service Manager for Research Computing Consulting, UCB
- Erik Mitchell, Associate CIO, UCB
- Laura Smart, Head of Digital Scholarship Services, UCI
Panel Moderators:
- Mary W. Elings, Assistant Director and Head of Technical Services, The Bancroft Library, UCB
- Carl Stahmer, Director of Data and Digital Scholarship, UCD
Panelist question responses will be posted here February 20, 2018.
What current use cases demonstrate a need for DS services?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
UCLA has had a DH minor and a DH graduate certificate since 2011. The library has been an active partner in working with faculty to support the curriculum by teaching in-class hands-on sessions, and connecting DH researchers with resources they need: primary sources, tools and technology as well as expertise. What we have seen is a remarkable increase in demand across disciplines and outside the curriculum.
Events such as UCLA’s Mapathon after the disaster in Puerto Rico, which was organized by librarians and held in UCLA’s Research Library, were powerful demonstrations of how web-based tools have evolved to the point that we can use our spaces and technology to pull-together communities of practice to around a cause, teach them how to use a tool, and put them to work. The library is no longer a storage house for books, but an active, dynamic, engaged collection of experts who work together with our users to connect them with information and information technologies to solve very real and urgent problems. We believe that technology is ripe, open, and accessible enough that without a lot of overhead, the library can empower researchers to actively engage with sophisticated tools utilizing digital resources.
Of course there are obstacles and issues, but from my perspective, the opportunities for librarians to become key partners in the research process, particularly Digital Scholarship, are no longer just buzz words for grant proposals or peer review, they are attainable goals.
For example, this quarter a team of librarians began offering–along with our campus partners–a workshop series: we called it Winter Wednesday Workshops. We built a curriculum that featured a “menu-like” assortment — from appetizers (presentation tools) to main courses such as R and Python. We have drawn over 500 participants to some 25 workshops. Participants came from all over campus: from Biochemistry to Classics to Electrical Engineering. We hoped for a small success but we were blown away by the demand.
Erik Mitchell, Associate CIO, UCB:
At UC Berkeley we see a lot of work in “Digital Scholarship” as well as a lot of discussion about what this phrase means anymore. As always instruction and infrastructure support are useful but it can be difficult to anticipate and meet needs. One service approach in general that we have seen have value is the provision of ‘consulting’ services whose main goal is to connect people with the right campus service provider. I have seen some substantial and valuable interactions come from that service where the outcome is less about providing access to a tool or expertise and more about helping researchers think through issues and understand their options better.
Another area where we have focused is on deep collaborative partnerships. Project Irene, Philobiblion, and the West African Arabic Manuscript System are all examples where the library has taken an active development, leadership and/or curation role. Of course each of the projects come with their own funding, sustainability and logistical issues and we have not taken them on lightly.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
There are many! Researchers create digital projects; undergraduates increasingly work with big data; humanities graduate students in the humanities seek exposure to and training in digital methods; liaison librarians are enthusiastic about learning new skills of value to the departments they serve; campus digital initiatives require library and archives expertise; open access creates opportunities for new publishing models and education. Here I’ll focus on just one use case: materials for text analysis.
Researchers wish for bulk download of text corpora in non-proprietary, machine-readable formats for computational analysis. Additionally, these kinds of “research ready” collections (or datasets) are increasingly desired for digital scholarship courses teaching computational methods. So, they have both research and pedagogical value. Providing access to such corpora aligns perfectly with the traditional mission of libraries and archives to provide materials for research; in this case, libraries and archives can make our archival collections accessible as data, negotiate database licences that allow for web scraping or data delivery, and train researchers on building and sharing corpora while respecting copyright and usage rights. At UC Berkeley, we are tackling all of these strategies.
Laura Smart, Head of Digital Scholarship Services, UCI:
The core difficulty in our field is answering the question, “what is digital scholarship?” The definition is often conflated with digital collections, digital humanities, scholarly communication, and data management & curation. Our use cases vary widely at UCI because our model of Digital Scholarship Services covers all of these areas. Some examples:
- • humanists with collections of 3D images and software looking to collaborate on grants to create access & preservation mechanisms as we currently have no appropriate repository to support this type of content
- • biomedical researchers managing grant compliance for data management and preservation
- • faculty authors looking for support funding open access publishing fees,
- • academic departments wanting to digitize materials and create online collections in support of a new graduate program
- • instructors looking to use Omeka to use digital collections in teaching multi-modal writing
- • graduate students seeking help with non-traditional dissertations with multi-media and GIS components
- • graduate student seeking help with organizing lab data as part of research assistant job
- • a small cohort of non-UC, Orange County organizations collecting OC data
What are the demographics of the patrons we are serving?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
We are seeing increased diversity among students doing digital research. Our Digital Research StartUp Program (DResSUP) has applicants from interdisciplinary programs: Chicano/Chicana Studies, Ethnic Studies, Global Studies, Urban Planning, and new programs such as the Master of Social Sciences that recruit a diverse group of students. We are working with these students on identifying gaps/erasures in our collections, and we are especially pleased when they connect with local communities, archives, and museums to seek out resources and, in some cases, they are able to bring back digitized materials as well as present their work to these communities.
Erik Mitchell, Associate CIO, UCB:
Across campus different programs tend to serve different patron groups. The D-Lab primarily serves graduate students but I see them connecting more broadly than that. The library intends to serve everyone and from what I can see that is true, although I think we have more contact time with students than faculty. For the consulting services – we often get someone who does not already have a campus network to draw on – I think that is a valuable place to focus our attention.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
Our patrons include faculty, visiting scholars and postdocs, graduate, undergraduate students, and staff (especially librarians and archivists!) and they represent a wide range of disciplines. In my experience, they are curious and enthusiastic, and they have a desire to dive in and create. They enjoy hands-on work and seek to understand what’s going on “under the hood” of a particular technology, and while some prefer independence, most favor working in teams or developing their independent projects within a community of practice.
It can be difficult to tailor services to so many different groups, and libraries may need to weigh local gaps and needs in order to strategically develop programs and resources. While digital scholarship needs may have some consistency across disciplines, you catch scholars’ attention best when you speak their language. Are your science faculty being well served by department-funded IT staff? Perhaps invest your attention elsewhere. Do you have a centralized DH program that supports high-level DH research? Focus on beginners. You don’t have to tackle everything at once.-Stacy Reardon
Laura Smart, Head of Digital Scholarship Services, UCI:
Graduate students and faculty constitute the majority of those using our Digital Scholarship Services with occasional use by campus staff such as coordinators of specific programs (international students, post-docs, summer undergraduate research programs, Office of Research, etc.).
Who are we missing?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
We know that students have not looked to the library for the types of workshops we are now offering. We’ve had to work really hard on getting the word out — through direct emails, using social media, etc. We had to learn basic things, such as, students may sign up, but they expect reminders, confirmations, etc. We’ve used social media, we’ve had to figure out branding, we’ve learned to be very humble about our lack of promotional abilities and how much work it takes to change preconceived notions.
To be honest, we miss people because the UCLA library lacked proactive ways of assessing demand. We track usage statistics for traditional library resources and in-course instruction, but we didn’t have a way to manage sign-up sheets and email addresses of participants. Rather than waiting for students to tell us what they want or didn’t care about, we have been sending out quick surveys (through campus mail) in order to ask whether grad students want library workshops on R, Python, Visualization, Statistical Tools, etc — and then we go find instructors among our partners willing to teach those things.
Sometimes, in our eagerness to showcase how technically savvy we are, we forget to offer the basics; Zotero and spreadsheets were very popular workshops and can be springboards to other offerings on our menu.
We are thriving in the high-touch, deeply engaged, incubator style environment (DResSUP) that we’ve created. It is small scale; we work with about 6-8 graduate students each summer.
We wish we had funding to provide scholarships to reach students who need financial assistance to stay in Los Angeles during the summer. We want the best of both worlds: maintain our high-touch approach, but amplify our impact. The workshops during the academic year seem to be an effective way of doing that. Our grad students return to teach workshops, share their projects, help us to recruit the next summer’s participants. What we hope we’ve created is an ecosystem for a digital scholarship community of practice with the library as the center of activity. DResSUP is our way of developing deeper connections and engagements (with users and colleagues) that then inform and help sustain our workshops, events, and consultations.
Erik Mitchell, Associate CIO, UCB:
Perhaps this is not a ‘who’ but a ‘how.’ I think it is increasingly difficult to provide DS services that meet a specific research need – especially when we focus on tools. The library has made some guesses in the last few years about what kind of research tools are needed and ultimately I think the landscape of DS research tools is too diverse for the library to support in a meaningful way. This is not necessarily the case with data where we have seen good adoption of a data licensing program. That is one area where the library has a unique place on campus (e.g. expertise in licensing, understanding of data acquisition, access and preservation) and funding to secure access to data. At the same time – this service really only meets the needs of a limited number of researchers and is only as useful as our outreach efforts.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
At large research universities, digital scholarship services may focus on the advanced projects of faculty and graduate students, which can mean undergraduates get less attention than they should. I think this is changing. We can learn a lot from small liberal arts programs, whose sole focus is undergraduate education, and where undergraduates are treated as co-researchers. Undergraduates can develop as scholars using DS methods in their own right, especially at the thesis stage; they can be valuable participants in digital projects; and they can serve as guides to digital methods for their peers. In the last year at UC Berkeley, we’ve seen an explosion of interest by and in undergraduates as digital scholarship researchers. Undergraduates founded a student association for digital humanities; undergraduate research and education were one important focus of the 2017 DH Faire (a weeklong symposium on DH at Berkeley); the Library is piloting an undergraduate digital scholarship consulting service and fellowship program; and the campus is launching its inaugural session of the DH Summer Minor in 2018.
Laura Smart, Head of Digital Scholarship Services, UCI:
We rarely serve undergraduates directly. There is room to successfully leverage support of digital collections and “literacies” (digital and data) within undergraduate curricula as a means to educate faculty and graduates about librarian expertise in these areas. Our organization largely allocates responsibility for undergraduate work to the liaison Page 2 of 3 librarians within our Education & Outreach and Reference departments. “Data literacy” and “data librarianship” are complicated turf internally due to longstanding perceptions that some liaison librarians have regarding undergraduate students and reference interactions involving the “obtaining and using data” parts of the data research lifecycle. The cultural shift necessary to integrate these services is ongoing.-Laura Smart
How are digital outputs changing our collection and preservation strategies and what changes should we make in the future?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
My vision of the future is that libraries should be flipped: that our front-end should be primarily collections of experts (librarians and library staff who normally work behind the scenes); the middleware is the methods, tools, and workflows, and the back-end is the materials and storage.
I would like to see librarians focus more on designing (collecting/sharing) research workflows: we are the guides to the iterative process of gathering materials, processing those, analyzing and interpreting, and then an output or sharing. I think researchers are interested in collecting their own materials (web-archiving, web-scraping, Twitter data, etc.)–and there are really great opportunities for librarians and researchers to work together throughout the process. If we felt less threatened by the fact that users want to do their own collecting, and were a bit more practiced in drawing connections from their needs to our expertise–in collection development and management, metadata, standards, workflows, etc–we would be more confident in asserting and promoting that expertise.
Erik Mitchell, Associate CIO, UCB:
Our patron-facing strategy to date has been to use DASH as a data publishing tool – we have seen limited adoption of that platform and so far it has been a good “base building” service to provide. In the last year or so we have thought a lot about the right investment needed to help us manage all of our digital assets. As you can imagine there is not one solution – protected or highly personal data needs one solution while highly redundant data needs another solution. We are keeping our eye on cloud services like Box and Google Drive to understand what role they might play in digital preservation in the future. From what we can see these tools provide rather robust data storage and preservation services but we need to make sure we make the right use of them.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
There is no question that the growth of digitized and born digital materials is creating challenges for both library collections and for archives in terms of how much we can collect (or provide access to) and how to preserve it. Here, I’ll focus specifically on the digital outputs of the scholars on our campuses. Even traditional research products like dissertations bring new complications in the digital age, as libraries build hosting infrastructure rather than bookshelves to house them, and newly minted PhDs turn to librarians for guidance on open access, copyright, and how the digital availability of dissertations may (or may not) affect future book deals.
Additionally, researchers across all fields increasingly recognize the value of making their datasets available for reuse. Libraries are well positioned to provide research data management services and data repository solutions that facilitate transparency and reproducibility, and in the UCs we are fortunate to have Merritt/DASH as a resource. Finally, researchers are developing multimodal projects that test the library’s ability to curate and preserve. Just as we collect traditional research outputs like monographs and journal articles, so too it makes sense for us to ensure digital research outputs are available for future generations of scholars. This is easier said than done, of course, since all born digital objects become obsolescent at a greater speed, and libraries may be unable to support both the digital object and the digital environment that sustains it, but we can strategize through a combination of advance planning, cataloging, documentation, researcher interviews, data collection, and web archiving services. One project we are working on at UC Berkeley is to promote the institution’s Archive-It service to researchers.
Laura Smart, Head of Digital Scholarship Services, UCI:
I don’t yet see wide scale change in collection and preservation strategies based on digital outputs. Some of our archivists and librarians have dabbled in web archiving, but our program is small. We’ve focused on some key regional government web sites as part of our regional data collection and select Orange County web sites as part of the Orange County and Southeast Asian Archive. Special Collections and Archives regularly digitize materials and consider preservation of born-digital collections they acquire. Bibliographers are resistant to helping their faculty shepherd digital projects citing being overwhelmed with traditional job responsibilities. Future changes we should consider obviously include work on technical and preservation related challenges. Research questions abound regarding topics like tracking provenance of data, integrating repository deposit into research work flow, etc. Before we consider that work, however, we have to cultivate a change in mind-set where librarians are willing to collect researcher output or manage digital collections as data. Programs like Data Rescue inculcate the notion that there is digital content in the world, that our researchers are using, which is very much worth collecting and stewarding.-
What additional or re-deployed resources and labor will be required to provide necessary services?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
What we’ve seen is that our metadata librarians, who are not typically considered Public Service librarians, are our strongest advocates. When users have a research project that touches on all aspects of the library, their view of the library’s roles in that process transforms. We believe that engaging these highly trained experts with early-career researchers builds community amongst librarians and contributes to a scholarly climate that makes professional archive, library, and museum work, cultural heritage work, valuable, accessible to the public, and visible.
Erik Mitchell, Associate CIO, UCB:
One benefit of the “consulting service” approach to providing DS services is that we can scale these services out by working with our librarians and departmental liaisons. That seems feasible in that we have someone focused on each department/school whose job it is to understand the scholarship occuring in that school and evaluate how the library might be of help.
There are other technology-focused services (e.g. access to desktop computers, analytic software, virtual desktops, HPC resources) as well as data-focused services (e.g. data acquisition, storage and preservation) which the library is investing in specifically. Recently we commissioned a “computing survey” that sought to understand the scope and impact of computing services in our libraries. The survey found (among other things) that there is still strong demand for simple machine and application access. At the same time – providing that service comes at a considerable cost which also prevents us from exploring and investing in more cutting edge services.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
Staffing needs will depend on what kind of digital scholarship model a library adopts and how significant an investment the library wishes to make. For example, for a digital scholarship model based around teaching, liaison and instructional librarians might develop digital scholarship training to complement their information literacy instruction, and subject specialists could work to develop knowledge of digital scholarship methods being used in their discipline. Revised job titles and descriptions could reflect this growing expertise. My current position is an example of a subject specialist role that was restructured to include a digital scholarship component.
To test the waters and gather data for digital scholarship resources, interested staff might conduct lightweight pilots to justify further investment. Over time, contributions from existing staff may be feasible and even welcome. For example, can the metadata librarian be available for a couple of hours per week to collaborate on faculty projects? Might the film studies librarian hold office hours in a digital scholarship lab? Graduate and undergraduate peer mentorship can both scale support and provide a degree of disciplinary particularity not otherwise possible.
Still, libraries are already being tasked with doing more with less. If the library has ambitious goals and seeks to develop substantial programs, departments, spaces, or infrastructure, then additional or redeployed staff will be needed. Understaffed initiatives may have the obverse effect of alienating potential users. Official launches of a new service are valuable opportunities to advocate for the additional hiring lines needed to make an initiative successful and to raise funds through donations and grants. Collaborations with campus groups beyond the library can help curtail the amount of funding any one organization contributes.
Laura Smart, Head of Digital Scholarship Services, UCI:
“Additional” services are not feasible within our current environment of shrinking budget and attrition-based loss of FTE. Re-deployed resources and labor are therefore the only option to provide new digital scholarship based services. At UCI the working model is that public service librarians provide initial consultations and refer to Digital Scholarship when detailed expertise is needed. They are expected to participate in projects involving researchers within their liaison areas. In practice referrals have been immediate handoffs. It’s a significant challenge. Re-deployment to digital scholarship means reducing or stopping other work librarians are accustomed to, and very invested in, doing. On the national level, other organizations have been examining the role of liaison librarians in digital scholarship yet these efforts appear focused on scholarly communication and data management and less with digital collections projects or digital humanities. There are prominent exceptions (i.e. UCLA DRESSUP) of course.
Are current and imagined services sustainable compared to more traditional library services?
Zoe Borovsky, Librarian for Digital Research and Scholarship, UCLA:
We’ve built all of our programs on a shoestring budget using a “minimal-computing” approach. We realize the demands exceed our current supply, but we’re looking to more peer-to-peer training with graduate students as well as partnerships with the graduate division and undergraduate research center as ways to continue to build sustainable communities (dare I say “ecosystems”) of practice.
One of the main limitations we have encountered is low wages for student workers in the library; equivalent positions in campus research centers have academic status and include tuition, making it difficult for the library to recruit and retain talented students. The solution we prefer would be to establish partnerships with academic departments and centers in order to fund graduate student positions in the library that would include tuition and fees. We envision not just information studies students at the reference desk, but a cohort of students from various disciplines engaged in consultation with their peers and librarians on digital scholarship projects.
Obviously, this level of imagined services would require more investment from academic departments, who, I believe would benefit, not just from the services but also from being able to offer these types of positions to graduate students. I would like to believe it would be worth trying!
Erik Mitchell, Associate CIO, UCB:
One strategy we are pursuing for “imagined services” or services in development is to build a service based on a specific use case but gear the development to more generalized use cases. For example the West African Arabic Manuscript project uses a multilingual index and abstract platform – something that is commonly needed in a certain type of digital scholarship project. We have been working over the last year to secure funding to develop a more generalized platform that many scholars could use for their I/A and project needs. Ultimately – I am not convinced that such a strategy works for more than a few use cases at any given time but where it works it can be pretty powerful.
Another strategy we are working towards is to focus some of our digital scholarship services on a cost-recovery model that would allow us to scale up the service as needed. At the moment our imaging services unit is the only one that follows such a model but we can supplement that unit using patron-initiated requests. We have yet to figure out how to leverage that service across campus but are using it to build out a digital lifecycle program within UCB Library that seeks to make all sorts of analog/physical scholarship available in digital formats.
Stacy Reardon, Literatures and Digital Humanities Librarian, UCB:
In the long term, absolutely — libraries and archives will develop the infrastructure and staffing necessary to effectively meet the evolving research paradigms of our scholars. It’s the road ahead that can be complicated. The work is fun and exciting, but it’s also frustrating and sometimes exhausting. In our budget climate, the library doesn’t have bottomless resources, and it can be impossible to both provide the traditional services that are so essential to facilitating research on campus while also building new systems, investing in digital scholarship services and spaces, and hiring or redistributing existing staff. Too often digital scholarship work is not given the funding or staffing needed to succeed. Projects may be one-offs that don’t contribute to a larger service model or lead to a greater investment in infrastructure, and efforts outside the scope of specific departments may languish when individuals leave or move on to other things. The library would do best to extend into the digital arena the core services in which we are strongest: promote the library space as interdisciplinary hub for discovery; make our collections accessible for computational work; teach digital literacies in tandem with information literacy; and provide one-on-one, discipline-specific digital scholarship consultations to researchers.
Laura Smart, Head of Digital Scholarship Services, UCI:
It’s difficult to discuss sustainability when we struggle to launch pilots and obtain appropriate staffing. Executives cite commitment to digital scholarship as vital to 21st Century library services. Change is slow to occur and likely requires a stronger executive mandate to accommodate re-deployment and re-allocation. Sustainability is possible but only if we avoid failure to launch.