Catherine DeRose is the Program Manager for the Digital Humanities Lab at Yale University, where she consults on digital humanities projects, teaches workshops on data analysis and visualization, and directs the Digital Humanities Teaching Fellows program. She received her PhD in English from the University of Wisconsin-Madison.
Peter Leonard is the Director of the Digital Humanities Lab at Yale University. He received his BA in art history from the University of Chicago and his PhD in Scandinavian literature from the University of Washington. Before coming to Yale in 2013 as the first Librarian for Digital Humanities Research, he served as a postdoctoral researcher in text-mining at UCLA, supported by a Google Digital Humanities Research Award.
This is the source
A 1930s reading room at Yale University Library is the site of an architectural
transformation that seeks to make DH praxis visible in a collaborative, open setting.
What design and policy interventions lead to the best use of this central and
symbolic space? Ethnographic study, user-centered design and a focus on the
materiality of both physical and digital collections combine to suggest one pathway
for research libraries to support collaborative digital work in the humanities. In
this article, two digital humanities staff at Yale Library discuss the relationship
between inclusion v. separation, security v. transparency, and historicizing v.
two digital humanities staff at Yale Library discuss the relationship
between inclusion v. separation, security v. transparency, and historicizing v.
A historic reading room from 1931 is the site of one of the newest scholarly support
environments at Yale University: The Franke Family Digital Humanities Laboratory.
While neither the first renovation of the space — the room shifted from reserve books
to periodicals in the 1970s — nor the first home of the lab, which relocated from a
room requiring a campus ID for access to a publicly accessible, much larger space —
the transition of a reading room to a digital humanities (DH) laboratory represented
a new kind of commitment from the Library that required structural, aesthetic, and
programmatic transformations. The goal was to create a collaborative, open workspace
where the practice of digital humanities would be visible, accessible, and supported
within Sterling Memorial Library, the so-named heart of the
university
humanities infrastructure,
with a focus on how
location, materials, and services could be designed to support DH research and
training within libraries.
The transformation of this particular room, and the services found inside it, are
part of a larger process of libraries formalizing their support for digital
humanities in response to local needs and resourcing
In the 2019
more libraries are recognizing that they may want to invest resources, including staff expertise and time, technology infrastructure such as repositories, and physical spaces such as digital scholarship centers or labs, to make a more formal commitment by the library to this type of scholarship
At Yale, the Digital Humanities Lab (DHLab) has always been physically and organizationally located within the Library, tasked from the beginning with supporting scholarship across the humanities broadly considered. In addition to departmental diversity, the audience for DH support has included a broad set of clients, from Yale College students, graduate students, faculty, curators, and librarians to occasional collaborations with external partners. This range of disciplinary and professional perspectives presented unique opportunities and challenges for designing a space that would be inspiring, inclusive, and functional.
Moving DH labs into libraries invites a rethinking of library spaces that were
originally designed for oftentimes siloed engagements. Completed in 1931, Sterling
Memorial Library contains nearly 450,000 square feet of space to support learning and
research. Positioned immediately to the left of the main entrance, the Reserve Book
Room (Figure 2) was an integral part of architect James Gamble Rogers’ design. The
room and its support areas total over 4,000 square feet and were designed to hold
10,000 volumes on wooden shelves that flank the northwest and southeast walls. Large
expanses of leaded glass windows bring filtered natural light into the room, rising
to a ceiling of decorative plasterwork and chandeliers. The abundance of
craftsmanship is characteristic of Sterling Memorial Library as a whole, completed at
a time when the Great Depression enabled new campus construction nationwide at
relatively low cost. Yet the Reserve Room itself was described by contemporary
observers as a place of relative moderation: Inasmuch as the
Reserve Book Room is really a work room where students go for study,
wrote
alum Ellery Husted in the
its architectural treatment is more restrainedthan the other large reading rooms which filled the main floor of the new library
By the early 2010s, consensus had built on campus that the room could serve a more
prominent role in the future of research in the arts and humanities. With the
availability of other rooms within Sterling to hold print periodicals, the space
seemed ripe for reimagination. Discussions imagined a Digital Center for Arts and
Humanities,
modeled on the successful launch of the Center for Science and
Social Science Information (CSSSI) in a 1960s building in 2012
Indeed, the design and materiality of the Franke Room was naturally much different than the 1960s environment in which CSSSI was built. Instead of the basement of a mid-century building, the Franke Room contained wooden carvings, leaded glass, and plasterwork designed to evoke the early English Renaissance. Built-in bookcases lined the room up to a height of nine feet, transitioning to walls that were designed to appear uneven, as if finished by hand.
At the same time, the integration of the Franke Room into the architectural and material fabric of Sterling Memorial Library also offered advantages (Figure 3). In contrast to the disciplinary-specific reading rooms on higher floors (and behind access gates that require Yale IDs), the Franke Room is near the main entrance on the ground floor, open and accessible to all, making it a strong candidate for a large and welcoming space designated for the exploration of new ideas and methods.
As a design team of architects and library staff examined the room’s complex materials and physical adjacencies, several intriguing possibilities emerged. Given the room’s close proximity to the library’s physical collections, the reimagined space could position DH as contiguous with older traditions of humanities research rather than a radical break from it. The presence of a substantial print collection in the building could serve as a reminder that the root material for many digital projects is the print book or periodical. From a workflow perspective, this proximity could also streamline the research process: if the lab had equipment available for digitization, researchers could pull books from the stacks and bring them straight to the lab where they could transform them into machine-actionable data.
The original infrastructure of the room also contained large amounts of hidden
reserve book shelving behind a long, adjacent wall that was accessible only to
library staff. The design of the DHLab re-imagined this back-of-house
area,
opening it up to support collaborative meeting spaces and on-site offices for lab
staff. The 1930s ideal of hiding the mechanical aspects of scholarly support space
would give way to exposing this previously backstage area to the patrons in the main
room.
Planning for DH support at Yale coalesced in the early 2010s, ranging from
conferences
Whereas the sciences have long-established models for constructing lab spaces
outfitted to their research needs, the humanities have operated more in a tradition
of individual work. DH labs create opportunities for collaborative work that brings
together different sets of expertise and equipment to pursue humanistic inquiries at
varying scales. The challenge in designing such a space revolves, in part, around
striking the right balance between different needs and possibilities. For example,
should the room be a fully open workspace, or should it provide areas explicitly
marked out for different functions? Which is more likely to spur activity: a space of
unstructured possibility, or one that actively suggests particular modes of
engagement? Is a computer with a large monitor actively displaying an interactive
visualization more, or less welcoming, than an open table? In describing Penn
Libraries’ integration of DH, Anu Vedantham and Dot Porter underscore that [s]paces designed to encourage brainstorming and discussion need to
look, feel, and function differently from those designed to support deep reading,
note taking, writing, or presentation practice
A workspace unencumbered by machines affords different kinds of engagement with a
space and the people inside it than does one filled with monitors. The computer
lab
concept, with rows of fixed workstations, seemed to the design team to
belong to an older model, which was confirmed by site visits to newly designed spaces
that were more flexible in their layout, with some equipment balanced by open
workspaces (Northeastern, Brown, and others). In their 2015 survey
noted variation in the kind offacilitiesthese organizations occupied; collaborative space seemed to be more important than top-notch hardware
To acknowledge, anticipate, and address, the specific DH needs on our campus, the
Library engaged design anthropologist Nancy Fried Foster, formerly the senior
anthropologist at Ithaka S+R, for an information-gathering phase during which she
helped collect input from the Yale community. With Fried Foster, the design team
conducted interviews during 2016 with current clients of the DHLab, as well as those
who had not used such services in the past. The goal of including researchers who did
not identify as digital humanists was to ensure that the design team did not
overfit
the service portfolio and architectural design to those who had
already availed themselves of DH support services on campus. Fried Foster describes
the user-centered design process as one that begins with an
understanding of work practices and obstacles and then designs solutions to
support workflows and address identified problems. It is inclusive and
participatory, and is intended to improve outcomes by building a foundation of
valid, up-to-date information about the community for use by designers and
architects
The design workshops were particularly illuminating for seeing how researchers imagined engaging with the new space. The first workshop asked participants to draw their ideal spaces, including what technology they would like to see in the room and what sort of activities should be supported. The outputs from the first workshop became the basis for the second workshop, which took the list of imagined equipment and activities that might take place in the space (such as designated workshop areas, staff offices, desktop computers, flexible workstations, reception desks) and printed them on separate strips of papers. Participants then arranged those pieces of paper around a space roughly scaled to the Franke Room. Each strip had multiple copies so that if, for instance, a participant thought there should be five desktop computers and two workshop areas, they could arrange the room accordingly. There was no requirement that participants use strips from all of the different resource or activity types.
Patterns from the sessions helped inform priorities for the renovation, including the
need for separate programmatic spaces for consultations, collaborative work, and
project exhibitions. Participants wanted to have ready access to DHLab staff, but
they recognized the need for staff to have occasional privacy as well. Large display
screens emerged as a significant request for both collaborative work and exhibitions,
which drove conversations among the design team toward the notion of a large-scale
data canvas
as a possibility for visualizations. Meanwhile, conventional
desktop machines were lower on the list of must-have equipment: most students,
faculty, and staff preferred working from their own laptops, but they acknowledged
that a few high-end desktop computers in the DHLab would be helpful for accessing
software and for more computationally intensive projects.
While the feedback received during the information-gathering phase was generally
consistent, Fried Foster’s summative report
During the time that design team was gathering requirements and beginning the architectural rendering process, the newly-formed DHLab — consisting of a Director (formerly the Digital Humanities Librarian) and the newly hired Engagement and Outreach Manager, User Experience Designer, and Digital Humanities Developer — began work in the repurposed rooms on the third floor. The constraints and opportunities of this temporary space led to several lessons that would, coupled with the design team’s assessment, inform the eventual design of the Franke Family Digital Humanities Laboratory. In particular, the team learned that placing its reference print collection of DH books on the old wooden shelves started innumerable conversations with curious visitors as well as clients who made appointments to meet with staff or other collaborators. As every book was placed on a metal stand that displayed its cover to onlookers, the breadth and diversity of scholarly publications in titles as
Big Datamoment:
Serving as a focal point for — and for many, an introduction to — DH on campus, it is especially important for labs to be inviting, inclusive spaces, welcoming to long-standing DH practitioners, the curious but unsure, and everyone in between. In conceptualizing the layout of the space (Figure 4), the team advocated for modular designs that would support multiple kinds of DH engagements. DH is a robust, rapidly developing area of research and teaching, and the goal was to design a space that could respond to new needs and forms of collaboration that might surface. Nearly half of the DHLab would thus consist of an open workspace occupied only by wheeled tables and chairs. The tables, which can be linked together at the ends, can be easily reconfigured into pod formations, long rows, or U-shapes to support workshops, presentations, working groups, collaborations, and individuals working on laptops or reading books. Walls would feature alternating panels that incorporated both print book collections and digital visualization monitors.
One service point with a track record of inspiring initial conversations was the print book collection that had lined the shelves of the temporary space. The team wanted to expand and highlight this collection in the new, larger room — gesturing towards the original function of the space. With new support from central Collection Development funds, the books became a consistently growing resource — as well as a dramatic visual characteristic of the space. The collection was ingested into the central Yale Library catalog thanks to the work of the Technical Services team, who adjusted their workflow to accommodate call letter labels that would not obscure cover designs and who added custom bookplates designed by the User Experience specialist in the DHLab. In the higher-profile location along the walls of the renovated room, the books encourage patrons to take up theoretical debates in the field, develop specific coding skills for their research, or read up on DH applications within specific disciplines.
Thematically organized, the books provide a snapshot view of the four DH areas the DHLab is most equipped to support (text, image, network, and spatial analysis), in addition to highlighting several books on data visualization, web development, and the DH field at large. The goal for the collection is for it to be a resource to which scholars can turn for instruction and inspiration. With the increasing uptake in e-resources and library services where books can be pulled from the shelves for patrons, there is less browsing — and by extension, fewer unanticipated connections being made — in the stacks. To stimulate engagement with the DHLab’s print collection, the architects designed slanted shelves that could display books covers facing out, as the DHLab team learned from the temporary space that researchers coming to the lab for a table to work at might also find themselves drawn toward a book on telling stories with data visualizations.
A challenge but also affordance of an open workspace is the lack of sound barriers. While the sound carryover can be distracting for work that requires intense concentration, it also opens up the possibility for serendipitous connections. Organic collaborations, as the DHLab witnessed in the temporary space, often form in such communal spaces, where a group working on a given project realizes they had something to offer a group working on a different effort and vice versa because of overheard conversations. In this way, the team hopes that the opportunity for collaborations encouraged by the open nature of the room outweigh any disadvantage from sound travel.
Despite the attention paid to collaboration in library design, a significant
fraction of DH activity occurs as an individual activity. This consideration
emerged during the information-gathering phase, with participants requesting a
space where they might, as Fried Foster described it, work alone together
Sterling Memorial Library contains a multitude of dedicated quiet spaces for
solitary research; what it has fewer of are designated collaborative discussion
spaces where researchers can analyze, debate, and discover out loud. Fried
Foster’s report reflects this need as well, with students, faculty, and staff
expressing the expectation that the DHLab would be a space conducive to interaction and the sharing of ideas and information. They
envision an environment in which seeing others, working in company, and finding
opportunities to talk will foster serendipitous connections
But an effective discussion space requires more than just encouragement to talk out loud. Certain kinds of furniture are more conductive than others to grouping around a text or computer screen. The architects, who were provided with a copy of Fried Foster’s 2016 report and who met regularly with DHLab and Library staff, were mindful of the kinds of engagements that might take place in different quadrants of the space when they made their furniture recommendations. Ergonomic considerations, such as whether chairs in front of computers should have armrests or not, together with a focus on accessibility for patrons using mobility devices, informed the choice of furniture and surfaces in the space.
Teaching workshops in an open space rather than a computer lab offers serious advantages as well as disadvantages that must be taken into account when determining the layout and equipment for a DH lab. To begin with the disadvantages, participants are required to bring a computer with them, and not everyone has access to a personal laptop. Many libraries have laptop checkouts available to students, faculty, and staff; however, they do not usually grant administrative access, meaning participants would be unable to download any software. Tablets can also be insufficient substitutes for laptops when it comes to certain software. Even for the participants who can bring laptops over which they have administrative control, other challenges emerge. Personal laptops can be slow, low on memory, limited in storage space, and/or suffer from low battery capacity, which can pose a problem depending on where and how many outlets are in the space.
Conversely, these latter disadvantages point to some of the benefits of participants working from their own machines. Since installations can be quirky and setting up software environments daunting and confusing, it can be helpful for participants to walk through the process with an instructor. If they’re able to get everything up and running during the workshop, they can then leave it and continue practicing without having to learn first how to install everything from scratch by themselves. This is why Software Carpentry, of which Yale Library is affiliated through the New England Software Carpentry Library Consortium (NESCLiC), operates under the expectation that participants will work from their own machines.
Hosting workshops in an open space also yields a more collaborative environment. Rather than being hidden by monitors, participants can see one another, along with the instructor. Additionally, anyone working in other sections of the DHLab can also see and hear the workshop. While workshop registration is capped at a manageable number, the open space provides an opportunity for those who did not register to listen in and, if they are so inclined, to follow along on their own laptops.
Efforts to ensure that the work
of DH had a place in the room led to
discussion of how to best highlight the specialized equipment and activities that
characterizes some (but not all) of this newer form of humanities research. Fried
Foster’s final assessment report captures that much of the on-campus excitement
for a dedicated space for DH emerged from the hope that the space might catalyze
new work by making DH research visible. In the report, Fried Foster notes that
researchers want the set-up of the space to build excitement
and stimulate new work; [students, faculty, and staff] see this supported
through exhibits, displays, books, and perhaps even a museum, of projects,
artifacts, images, and sounds. Displays and exhibits should be visible
throughout the space and include screens
In some cases, the specialized nature DH equipment created conditions requiring a secured space. The glass Special Projects Cube seeks to fulfill that requirement while still making DH practice visible. One half of the cube includes equipment for transforming print materials into machine-actionable corpora, while the other half includes machines geared toward high-performance computing. Researchers passing through one end of the lab to the other can see the labor involved in running a collection of texts through optical character recognition (OCR) software or generating scripts to process thousands of images.
Scholars have been duplicating library material at self-service copiers for
decades, subject to the fair use provisions of US copyright laws. But advanced
text-mining projects call for a more comprehensive approach: the transformation of
physical collections into digitally-actionable research objects. For this reason,
the Special Projects Cube includes book- and microfilm scanners to aid humanists
in the creation of textual corpora. Crucially, this research digitization
practice is not about the reproduction or preservation of an original material
object: in most cases, the scanned images are discarded as soon as the textual
extraction process is complete, leaving only a transformed digital text.
The bound volume is of course not the only form of raw material: Sterling Memorial Library’s microfilm collections is large, but often under-utilized due to difficulties in handling and extracting the thousands of images often present on each reel. The DHLab, in conjunction with other library departments, installed a microfilm scanner capable of extracting documents from microfilm reels more efficiently. The resulting images, segmented by a machine vision algorithm, can then processed through OCR software installed on the DHLab’s machines to create a textual corpus for later analysis.
One of the aesthetic challenges that emerged during the information-gathering
phase was how to incorporate old furnishings with new technologies. The Yale
researchers who participated in our process indicated a preference for modern
equipment alongside a desire to retain a feel for the a library setting.
The design team knew that preserving the character of the room would thus be
essential. The architects did make two changes to bookcases: re-orienting the
shelves to display book covers facing outwards, and installing digital screens
carefully-matched to the scale and structure of the bookcases, flush with the wall
envelope. The result was an updated look for the room without changing the
fundamental geometry or finishes.
The orientation of the room along a long axis is reinforced by the way that patrons enter at one end: the foyer has a much lower ceiling that creates a sense of compression. Proceeding into the room itself draws the eyes forward to large windows on the far wall. Subdividing the space with any materials or objects that blocked this axial view would fundamentally change the room and the experience of entering it. A key design priority was thus to preserve as much of the view corridor throughout the room as possible. This dictated the dimensions and placement of the Special Projects Cube, which is located offset from the center of the space but directly in the line of view of a patron entering the room (Figure 3). The surfaces of the cube are made of glass, which allows the transmission of light and motion from all directions. In addition, the intervention of a semi-transparent object in the middle of the room heightens the visual drama of the space, as viewers have both a filtered view of the length of the room, and the people working in the cube.
The fact that DHLab staff, in the temporary space on the third floor, worked in an
open environment right beside computer workstations meant we experienced auditory
interference from overheard conversations, along with a constant sense of being
on call
for impromptu tech support. On the one hand, this arrangement
provided valuable insights into the projects people were taking on and the
technical or methodological difficulties they were encountering. On the other
hand, it meant that professional staff might be interrupted at any moment and
pulled away from other support efforts that required sustained concentration.
For the renovation, the DHLab staff wanted to retain awareness of the kinds of work clients were doing in the space, along with being visible ourselves so that we could provide support, but we (as well as the faculty and students Fried Foster interviewed) also recognized the need to work quietly without interruption at times. The final design result was to locate the staff offices along the periphery of the main space, with a glass wall on the one side that faces the space, thereby enabling staff to see out and researchers to see in. An open door signifies staff are available for questions, while a closed door asks that you return later. The signs on all offices display the schedule for the DHLab’s open office hours, a regular time during which staff are reliably available for drop-in consultations (researchers who cannot meet during office hours can schedule consultations for other times).
Implementing this design involved creating several openings in a structural wall to transform what was the former bookstack annex into a space for human specialists. DHLab staff worked alongside architects to optimize the offices for DH workflows. For instance, the DHLab’s User Experience Designer and Developer share an inner sliding door they can push back so that they can, on occasion, turn their two offices into one, allowing them to more easily converse during production sprints. The repurposing of the annex created designated, visible spaces for expertise and support that also provided DHLab staff with mechanisms for signaling availability.
The meeting room, a rectangular space that occupies part of the former annex and is therefore also adjacent to the main open lab space, provides DHLab staff and collaborators with a sound-controlled location for presentations, meetings, and smaller classes. Display screens are affixed to each of the shorter walls in the room to facilitate presentations. On one of the longer walls, there is a magnetic, glass whiteboard to help with sketching ideas and designs, while the opposite wall acknowledges Yale’s early interest in humanities computing. A poster of Father Roberto Busa working on a teletype at Yale in 1956 is affixed to one end of the wall, while a poster of the cover page to the proceedings of the
Although DH labs can serve as effective hubs on campuses for computational work, they are often not the only places where digital work occurs. Studying the larger ecosystem of where such work takes place can help to identify opportunities for cross-unit collaborations, whether that take the form of shared expertise or equipment. At Yale, two centers that predate the DHLab — the Center for Engineering Innovation and Design (CEID) and the Center for Collaborative Arts and Media (CCAM) — have 3D printers available for researchers to use, along with training sessions to help people get started. When it was time to make a decision about what to include in the DHLab, the team decided that the strength of other campus units in 3D printing meant that we could refer clients to those existing service points (and their accompanying expert staff). This freed up space in the Special Projects Cube for other kinds of equipment less readily accessed by humanists on campus.
To bring the expertise of subject librarians and curators — those most familiar with the promise and idiosyncrasies of the Library’s holdings — into DH conversations, the DHLab has designated four touchdown spaces where Library staff can meet with researchers or demonstrate projects of interest. Located in the four corners of the main space, the touchdown desks signal specialized support. For instance, the Geographic Information Specialist (GIS) Librarian, whose office is located in CSSSI, holds weekly office hours in the DHLab. She meets with researchers at the corner desk that is located next to the GIS books in the space so that she may easily refer to texts in the collection that might be helpful resources. Even when she is not physically present in the DHLab, researchers may still find her contact information on the desk. Along with serving as touchdown spaces for library and museum professionals on campus, the corner desks also provide a workstation for visiting scholars who might be on campus to give a talk or workshop in the space.
The transformation of a 1931 Reserve Book Reading Room into the Franke Family Digital Humanities Laboratory involved new ways of thinking about location, materiality, transparency, security, collaboration, and adjacency. The challenge of adaptive reuse of a historic (and historicizing) space is undoubtedly different from the process of renovating a newer, less decorative facility — or constructing a completely new room. The diverse team that came together to accomplish this — student and faculty participants, library and IT professionals, external architects, and university facilities specialists — were well aware of the significance of this room to Sterling Memorial Library’s design and history, as well as the scholarly practice that is emerging around DH on campus. The various interventions in structure, materials, function, and organization of the room to transform it into the Franke Family Digital Humanities Laboratory involved both empirical evidence, discussions with campus stakeholders, and educated guesses about what DH scholarship might evolve towards in an era when interdisciplinarity is increasingly inflected by the algorithm and the dataset. Yale’s deep commitment to the Humanities, and its Library’s scholarly support for the same, has always been expressed in world-class physical holdings and subject specialist expertise. The design of the Franke Family Digital Humanities Laboratory is an effort to reflect that same commitment, and that same support, in a new lab embedded within a historic room.
Heart of the UniversityTurns 75