This case study discusses a project that deals directly with building long-term sustainability into digital projects, with particular attention to the socio-cultural challenges of the project.
By Alison Langmead, Clinical Associate Professor and Director, Visual Media Workshop, University of Pittsburgh
Problem and Context Statement
The Socio-Technical Sustainability Roadmap project is one piece of a larger, sustainability-focused research agenda housed within the Visual Media Workshop (VMW)—the digital humanities space that I direct at the University of Pittsburgh. The VMW is located in the room that used to be the 35mm-slide library belonging to Pitt’s Department of History of Art and Architecture, and indeed when I began my job, the slides were still the main tenants. But I inherited more than just tens-of-thousands of slides when I assumed responsibility for the old space. I also inherited, as steward, a number of digital projects that had been housed on a set of departmental servers associated with this facility. This was back in 2009, and at that time, University policy had recently changed to disallow individual departments from running their own web servers, and therefore part of my immediate responsibility became finding new homes for the the subset of these digital projects whose original creators were still involved and still wished to sustain their work.
One of these projects was, and is, the website, Images of Medieval Art and Architecture, which was begun in mid-1994 by Professor Alison Stones and Jane Vadnal, a student-colleague of hers at the time. Vadnal, who had made herself familiar with the early World Wide Web, had come to Stones prior to Summer 1994 with the idea of scanning and posting Stones’ extensive research collection of 35mm slides online. For a number of personal and professional reasons, Stones jumped on the idea to engage with these new technologies and wanted to begin by posting her images of medieval manuscripts, as these objects were the focus of her own research. However, especially back in the mid-nineties, the libraries and museums that held these objects commanded steep prices for the use of their images and kept strict control over their licensing. So, instead, the team focused their work on Stones’ own research photography of medieval architecture (that is, the images of hers to which she owned the rights) and began posting numerous scans online, as well as some custom-designed pedagogical materials for use in the classroom. Over the years, the collection of images grew while the pedagogical materials, for the most part, receded from view.
Having been handed the responsibility for such an august, but relatively stable, project, I began to truly wonder what constituted my charge as steward. I had a personal attachment to Images of Medieval Art and Architecture, as I had known of the existence of the site almost from its very inception. My own doctoral studies were focused on 11th– and 12th-century Western European architecture, and I had always taken a “Humanities Computing” approach to my work, which included work with digital imagery and a fascination with the World Wide Web. By the time I arrived at Pitt, however, Alison Stones was heading off to retirement, and the others who had also been working on MedArt over the years, including Jane Vadnal, were moving on to other projects. Alison still wished to be involved in the work and had not fully and/or officially handed it over to me. And yet, the exigencies of my job title and the institutional infrastructure that surrounded me necessitated that I was responsible for its upkeep in many respects, especially its technical needs.
But what were the precise boundaries of my responsibilities? What should I be doing with the project moving forward? I knew that taking MedArt offline was out of the question, but was still not sure what the right path for my work on the project should be. These images—many of them scanned only to low, 1990s-era standards—were essentially plopped up on the Internet using simple HTML and were not of as much use in 2010 as they were in 1997. Indeed, it was something of a miracle that this historic site still existed. What was its fundamental purpose now? Surely better images could be found elsewhere? Should I extract all of MedArt’s images and put them into a new, fancy, indexed content management system? That seemed to miss the point. Should I leave the site the way it is and at least add a sitewide search feature?2
It wasn’t clear that this would increase the usability of the site in a way commensurate to the additional work, both in the present and on an ongoing basis, that it would entail.
There were a number of stakeholders prominent in my mind as we thought about this work, including (and perhaps especially) Alison Stones herself. She was still very much involved in MedArt’s persistence, and yet she was at a remove, having moved to France for her retirement. I was very much aware of the fact that I was interacting with, and making decisions about, a project that she had been contributing to for over 20 years and that was now undergoing a massive change in stewardship. I was also cognizant of the fact that we were thinking of MedArt from an outside, even historical, point-of-view that she very may well not have ever considered. Indeed, during our later interviews with Stones, we discovered that this was very much the case—she had not ever considered MedArt in this sort of historical light—but she quickly revealed that she was also thrilled to be able to come at this work again from a different angle, allowing her to see well-known, well-loved things in a new light. I am happy to report that Alison has become a prominent supporter of our stewardship work and I feel that her continued engagement with the project is one sign of our being able to successfully balance the needs of our past and present stakeholders successfully.
Jane Vadnal, the other founder of Images of Medieval Art and Architecture, is another prominent stakeholder in this work, but one who has remained elusive. Over the years of working with the site, both the VMW team and Alison Stones have repeatedly tried to be in contact with her to learn more about her involvement in the project and more about its technical framework, but she has not returned our phone calls or emails. Her absence is felt strongly. There are so many questions we would like to ask her. We respect her decision to stay at a distance, but also recognize that she very much remains a stakeholder in the past and present of MedArt.
Many of the members of the Department of History of Art and Architecture here at Pitt were also major stakeholders in this work, not only as participants in the project as it originally unfolded, but also as informants for our research. Both Veronica Gazdik and Linda Hicks, longstanding staff members of the department, were utterly invaluable to us as we learned more about the history of MedArt and worked through our decision-making process about how to sustain it. Indeed, when I announced at a recent (2018) departmental meeting that the site would be maintained in its current, although gracefully degrading state, many members of the department applauded and all were pleased to hear that Images of Medieval Art and Architecture would remain a part of our community.
Getting to the Goal
This happy conclusion, however, was to be reached by means of a deliberate process that involved numerous actors over numerous years. As with many projects that last over relatively long periods of time, not all of the participants were present throughout–we had a changing cast of characters. Indeed, the work that went into deciding how the VMW would go about sustaining MedArt began all the way back in Academic Year 2012-2013 when, Brian Beaton, a former colleague of mine from the School of Information Sciences, and I began thinking of ways to instantiate research projects that were focused on digital sustainability but that could also involve and help educate our students. We thought we might produce a “DH Sustainability Calculator” that would assist digital humanists in selecting technologies that best fit their sustainability goals. To this end, we put in for a National Endowment for the Humanities Office of Digital Humanities grant to support the idea. While this application was not funded, Brian and I continued to think about how we might proceed with this project on a smaller-scale. It was in Spring 2014 that we began to consider producing an academic case study focused on the sustainability needs of a long-standing digital humanities project using MedArt as the focus.
Brian left Pitt to move to Cal Poly that summer and his transition saw him leave this project, but I continued to think about the ways that MedArt could be used as a case study for research on ongoing digital humanities sustainability needs. In Fall 2014, I engaged in a series of conversations with Aisling Quigley, a PhD student at the School of Information Sciences at Pitt who was researching and working in the VMW for that academic year, and we discussed the sorts of ways that the preservation (and/or re-creation?) of this project could be approached. Having both been trained as archivists, Aisling and I decided to move forward by focusing on a very small-scale, user-focused research project that would assess the ways that current users of MedArt perceived the site. It is my recollection that, at this point, we did this work thinking more like archivists than active project creators. We were trying to gather information to inform our practical decision-making process about what to do with the project itself in its current, preservable form, more than to instantiate this first study as part of any larger research agenda. I feel that our custodial approach would help us find numerous avenues of inquiry over time, even as we realized that contemporary digital “custody” now always involves active project maintenance/creation.
After distributing an online survey and analyzing the results for this mini-project, we argued that the usability of the site seemed fundamental to its ongoing sustainability, and probably, given what we had read of the literature, to many digital humanities projects. To distribute these findings, we presented a poster at the 2015 iConference in California.3Aisling exhibited this research at the poster session, and during her time dutifully standing in front of our work, she was lucky enough to have a conversation with Josh Sternfeld, then a Program Officer for the National Endowment for the Humanities Office of Preservation and Access. During this chat, Josh mentioned that this was the sort of project that might match well with his office’s funding opportunities.
I do not know if I can emphasize enough how important this conversation with Josh was. Aisling and I were not sure if or how this project would continue, and we were not even necessarily looking for outside funding at this time. But with Josh’s encouragement, we began to imagine the ways that a more in-depth case study revolving around MedArt might be made into a larger-scale project that could benefit others who were confronting the same task of preserving and/or sustaining the earliest, web-based, digital humanities projects. For me, the notion of a “calculator” (the one that I had designed with Brian) had receded from view as the clear socio-technical burdens of sustaining digital humanities projects had come to prominence not only in our preliminary research but also, frankly, within the work being produced in my own lab.4 Not only was I experiencing semester-over-semester student turnover in the VMW that created a constant need for training and re-training on ongoing projects, but the institutional environment surrounding me was changing as well–first slowly, and then dramatically as the School of Information Sciences was transformed into the School of Computing and Information. I was constantly navigating my social environment even as the technical environment was serving as its own force for change.
So, instead of creating a “calculator,” creating a facilitated, face-to-face workshop seemed a more fitting approach to this problem, one that would allow practitioners to learn and engage with contemporary digital preservation practices in a context that suited their own needs and environments. It became clear to us that not all projects would have the same sustainability goals, and perhaps even more importantly, research by Trevor Muñoz, Bethany Nowviskie, Trevor Owens, and Geoffrey Rockwell had been showing that not all digital humanities projects had even been considering what their sustainability goals might be.
To this end, we submitted a grant application for a project we called “Sustaining MedArt: The Impact of Socio-Technical Factors on Digital Preservation Strategies” to the National Endowment for the Humanities Office of Preservation and Access and were, in the end, successful. This grant promised two outcomes: (1) a case-study-based research project that would investigate the socio-technical sustainability needs for the Images of Medieval Art and Architecture website and (2) a web-based tool called (at the time) the “Socio-Technical Digital Preservation Roadmap” that would distill our findings from the case study into a resource that others could use.
We began this work by focusing almost entirely on the case study, revealing the past and present development and uses of Images of Medieval Art and Architecture. We were very interested delving deeply into the ways that MedArt’s current usability might impact the site’s sustainability and preservation-worthiness. To this end, we organized a team of five researchers from both the School of Information Sciences and the Dietrich School of Arts and Sciences at Pitt and descended on the 51st International Congress of Medieval Studies in Kalamazoo, Michigan in May 2016 to talk to current and potential users of the site. There, we used semi-structured interviews to gather what would end up being over one hundred individual and group responses to the website from this international community of medievalists. We took every opportunity we could to speak with attendees about this project, using coffee breaks, chance encounters in the halls, and the infamous Kalamazoo wine hours. We employed iPads for the survey which allowed participants the opportunity to see and interact with the site on the spot—many times for the first time. After returning from Kalamazoo, we began focusing on a large-scale research initiative that pieced together the history of the site—including performing interviews with as many of the project’s creators as possible—hoping that the past goals of this project could also help inform our future decisions about MedArt’s sustainability plans. And, indeed, all of this work did eventually deeply inform our thinking.
At the end of this phase of the project, we concluded that MedArt had become a time capsule of sorts. The site’s dated appearance was not a disadvantage to the project, contrary to our initial assumptions. Its look-and-feel had become, in many ways, its very calling card. For many of the interviewees, the site was a testament to the way the web used to be, and a reminder of a different time and place. If sufficient funding were raised, a major overhaul of the site could certainly be accomplished by replacing the old scans with newer, more vibrant images, and replacing the hand-coded HTML with a new, modern content management system with all possible technological bells and whistles. Such changes might even be welcomed in some corners. But, for most, if Images of Medieval Art and Architecture were to become “just another” image database, MedArt would simply cease to be MedArt. Those users who were familiar with the site stated in their interviews that the idea of MedArt as MedArt was very important to them, and those users who were not familiar with it did not express a need for an additional image database made up of these images. From these perspectives, then, an overhaul would either be tantamount to taking a time-tested, historical site down for no real advantage or making changes that would not provide the community with something they needed. Change can, of course, create destruction as well as improvement.
The Creation of the Socio-Technical Sustainability Roadmap
Perhaps nine months into our research into the history of the MedArt project, the team began to feel ready to address the needs of the second deliverable promised to the NEH: the web-based resource that would take what we were learning from our case study and would provide a framework for guiding others through the process of sustainability planning. The Sustaining MedArt project team was finding that effective, ongoing project sustainability relied in our case study, above all, on effective, ongoing project management. This was the foundational principle that we knew we wanted to build into the “web-based resource” that we would eventually name the Socio-Technical Sustainability Roadmap (STSR). Aisling and I had also been doing work on design thinking and other face-to-face interactive pedagogies for other work in our lives and careers, and we decided to put that work to good use here. The thought of creating the framework for a multi-day workshop focused on the project management and sustainability strategies being revealed by Sustaining MedArt became our preferred way to move forward and we began to scaffold our thinking around it.
It was over the course of Academic Year 2016-2017, then, that the Sustaining MedArt team began discussing what exact form the STSR would truly take. Aisling and I had a long conversation about the sorts of interactivity that we wanted our audience/participants to have, and we very quickly found that the sustainability needs of the very resource we were creating needed to be front and center in our minds. This was not only because of all of the findings we were revealing during our research on MedArt, but also because we were well aware of our own environment–we knew that we could not predict how much time the VMW staff would have to devote to this project after the grant period ended. If we wanted to produce a web-based resource with numerous forms and a serious amount of back-end processing of user-generated/contributed text, we knew that we would need to sustain those “bells and whistles” for the long-term. We felt that our emphasis on designing sustainability into this project from the get-go was doubly important for the STSR not least because of the irony it would entail to create a tenuously sustainable project that was, itself, focused on digital sustainability!
In terms of the actual material form that the STSR would take, our work on MedArt demonstrated that the fewer technologies we used, and frankly, the older the technologies that we used, the slower things would degrade over time and the less direct support the technology would need. So, instead of whiz-bang forms and text manipulation, we decided to use technologies that were as close to plain HTML as we could. First, we compromised on the use of WordPress rather than true, hand-coded HTML. WordPress has been around for years and years and has robust export features should it ever be the case that the project folds. Moreover, we were already running a network install of WordPress on the VMW servers, and so by using this platform, we would not be increasing in the number of technologies that the lab needed to sustain. That is, we had already committed to supporting WordPress for the medium- to long-term so we were not adding sustainability requirements to the staff or budget.
Furthermore, since we knew that the participants in the workshop should walk away from their convening with something tangible in their hands, we knew that we wanted to create worksheets for them to fill out. PDF, as a solidly infrastructural format, could help us produce handouts, but many of the actual worksheets that we had in mind needed a more interactive structure. We considered using Google Sheets, as their online nature would befit the STSR’s orientation towards group work, but at the end of the day, we felt that using online spreadsheets was (a) too beholden to Google’s own profit motive and (b) not under our direct control as much as we would like for a project that should last as long as possible. We considered text documents and Word documents, but ended up with an age-old choice: Excel. Not only is this format utterly infrastructural in terms of format longevity, we also felt that the participants in the workshop would be much more likely to be familiar with the way that Excel itself works. Finally, the spreadsheet format would allow us to scaffold a small amount of data normalization of the sustainability information as well.
The bonafide production of the technical components of the STSR website occurred during Summer Term 2017. The team met and decided on the web address—one of the most time-consuming and frustrating parts of the creation of any web-based project, in my opinion (!)—and I got to work creating a new WordPress instance on the VMW’s servers. Over the next academic year, we set to work drafting and crafting the workshop’s modules, both in terms of their content and their sequence. Chelsea Gunn, another PhD student at the (now) School of Computing and Information at Pitt whose primary research focus is the ongoing preservation of digital personal records, joined the team in full-force during this moment and contributed greatly to the work.
Writing content for the STSR felt quite a bit like writing a large research report. We would diagram the flow of the workshop on the VMW’s white boards and then would attack each module’s main text and worksheets iteratively. I would say that we iterated over each module at least four times. Many changes were sparked by the ongoing testing that we were able to do. The first real test run of the STSR happened in Fall 2017 for the Art Tracks team at the Carnegie Museum of Art. We then ran the workshop on the VMW-based project, Itinera. In Spring 2018, we were also lucky enough to be welcomed to the Maryland Institute for Technology in the Humanities to run an instance of the STSR for a pair of their projects. This trip to MITH marked our final test for this phase of the project. The STSR as it is posted now at http://sustainingdh.net is in a fully-tested state—although we are very sure that the iterations have not concluded. The final report to the NEH is also finished, and so the current push to make good on our promises to this granting institution is complete.
The NEH also became a prominent stakeholder in our work, of course. Quite early on in the project, Josh Sternfeld left his position in the NEH Division of Preservation and Access and Sarah Lepinski took over his role. Both Josh and Sarah have been staunch supporters of our research and I cannot thank them enough for the time and energy they have put into offering advice and guidance along the way. As we move to the next possible phase of this work—that is taking the Socio-Technical Sustainability Roadmap around the country—both Sarah and Jennifer Serventi (from the Office of Digital Humanities) have continuously and gracefully offered so much of their time and expertise to us as we figure out our next steps. Talk to the funders! They are amazing people who wish to see research flourish.
Next Steps and Outcomes
As with most research projects housed in the VMW, the creation of the STSR was a highly collaborative event. It involved the work of many more students and colleagues than I have mentioned by name here—they are cited on the project, though, so go and read their names and look up their work!—and this is always one of the most exciting outcomes for me. I enjoy watching ideas that come from many different points-of-view piece together, and, importantly, I am also always interested to see the ways that some ideas stay very different. I truly believe in this project and know that its effectiveness does not come from any one idea that any one person has offered. Instead, it is an accretion of a number of approaches to the topic, made by many engaged and intelligent people, structured together to make a whole that is, by its diversity, necessarily greater than its parts.
Another fantastic outcome of this project is that I learned, through direct, felt experience, that projects do not always have to start with a BANG! They can accrete and grow organically as well. The success that the STSR has had rests on a series of conversations that took place over the space of years. It started small and snowballed. Now, there was also a significant amount of “right place, right time” luck, but not everything worth its while in the digital humanities necessarily starts with a massive press release and a large number of tweets.
As briefly noted above, we have applied for (and were successful in receiving!) further funding from the NEH, this time from the Office of Digital Humanities, to take the Socio-Technical Sustainability Roadmap around the country. We will offer the workshop in five different places over the next year, in all parts of the United States. If we had not been funded, however, we would have continued just as this project started…small. I remain willing to take this project to any site that wishes to make it happen, but the workshop is also designed to allow groups to facilitate their own convenings. I feel that the STSR will continue to gain traction—or, if not, will fade from view—because the team chooses to keep pushing it along step by step, introducing it to new communities and listening and responding to feedback. Will it go on forever? No, it will not. Right now, our sustainability plan for this project looks ahead only three years (for more on setting these types of sustainability timelines, please do see the STSR itself). But at the end of those three years, who knows what the project will look like and what needs it might need to fulfill?
In terms of the less rosy parts of the work, this particular project had its share of personality conflicts, some of which were deleterious to the project’s timeline and, even, to the human relationships involved. This is not always the way that collaborations in the VMW (or anywhere) work, but this time, there were a few disciplinary boundaries that were defended more strenuously than I have experienced in other collaborations. This sort of boundary creation almost always creates conflict when empathy and mutual respect cannot be maintained by all parties.5 For example, nowadays humanists are often still trained by scholars who prize independent scholarly research above all, and they instill their students with this idea even as collaboration becomes more normalized in certain corners of the academy. This often puts students in a bit of a bind as they have either to work against the grain of their own advisor’s expectations and/or come to realize that they prefer independent research as well. Social scientists can be more open to the value of collaborative research, but it is not always a sure thing that they are! Moreover, the importance of foundational “theory” to humanists is not always shared by other fields, while the idea of a proactively stated “methodology” is not always completely understood by humanists. Another complexity in this project centered around the nature of credit and collaborative work itself. For many academics who have not been trained to collaborate or to see the value of collaboration, it is very difficult to feel valued and valuable when the final product cannot be considered “one’s very own.”
If I were to do this project again, I would plan differently the way that the graduate students who are to be funded on the project were selected and trained. By default, I am much more likely to let graduate students work through problems independently and productively find their own way on a project. For myself, I detest being micromanaged, and would consider it a large professional failure on my part should I micromanage others. However, I have learned on this project that I do need to be more cognizant of the way that my own sense of pride in my work and my native management style may get in the way of what my student-colleagues sometimes need. This seems obvious as I write it now, but management that attends to detail can sometimes be what a student craves and requires in order to take the next step in their education. I feel nowadays that it is more important for me to proactively ask about and discover what the students’ current stumbling blocks are, and then to provide direct help by offering advice that is in concert with what I have learned from listening to the students’ needs both in content and tone. I was not trained to manage a collaboration in my PhD program (although I did take a management class in my MLIS program), and I have come to be of the strong opinion that managing academics is an art of its own that I am only now learning to handle with any degree of grace.
In terms of final takeaways, when push comes to shove, I know that if I feel that I am not making a difference, or contributing to a larger community, with my projects, I struggle to maintain my focus and determination. I wonder if this is something that others feel as well. I could not tell you what “making a difference” might mean to you, but I suspect you know what it might be. I know that we all do not live in a world where we can just do exactly what we wish to do whenever we want to do it—things are far more complicated than that and the power structures that surround us never give us complete freedom. But I do think that I gain solace in my most frustrated and stifled of times in the fact that the work that we do as humanists can truly make a difference, especially if we keep moving ahead together.
- During our investigations into the history of MedArt, it is worth noting that Pitt’s University Library System, at that very moment, changed content management systems causing more visible disruption to MedArt than any previous technological change had ever done. This break-down remains apparent on many of the pages devoted to Chartres Cathedral, such as http://www.medart.pitt.edu/image/france/chartres/chartres-cathedral/Architecture/Exterior/Diagram/Chartres-Exterior-Main.html. It is possible that some of this rupture may be repaired soon, but it is also possible that this is just one sign of the project’s graceful degradation.
- As it happens, during our research into MedArt’s history, we discovered that a sitewide search feature was something that Vadnal had been working on almost since the beginning of the project, but it was never successfully implemented.
- “Sustaining MedArt: Assessing the Persistence and Longevity of a Pioneering Digital Humanities Project.” Poster presentation with Aisling Quigley, iConference 2015, Newport Beach, California, March 2015. http://hdl.handle.net/2142/73731
- Projects like Itinera (http://itinera.pitt.edu) and Decomposing Bodies (http://sites.haa.pitt.edu/db) were showing their own sustainability needs and requirements at that time as well.
- On the importance of empathy and mutual respect in interdisciplinary collaborations, please see Tracey Berg-Fulton, Alison Langmead, Thomas Lombardi, David Newbury, and Christopher Nygren. “A Role-Based Model for Successful Collaboration in Digital Art History.” International Journal for Digital Art History 3 (2018): 152-80. https://doi.org/10.11588/dah.2018.3.34297