Rob Nelson and Making Digital Scholarship

One of the earliest conversations I recorded for the podcast was with  Dr. Robert (Rob) K. Nelson, the director of the Digital Scholarship Lab at the University of Richmond in Richmond Virginia.  I first met rob working on the project called The History Engine in 2007. At that time the Digital Scholarship Lab was not as well known as it would become to the broader public.  As the director of the Digital Scholarship Lab,  Rob has been at the forefront of some of the most dynamic projects linked to the field of Digital Humanities in the public sphere. The DSL has developed multiple visualization projects under the umbrella of the American Panorama (AP) project. AP is described as “a historical atlas of the United States for the twenty-first century” and it combines in-depth research with interactive mapping techniques. The maps on AP present data-rich visualizations that explore questions around redlining, migration, and electoral politics. As a result, the DSL has become a point of entry for many people learning about digital humanities.


American Panorama, Race, History, Mapping, Topic Modeling, Visualization

The Conversation

Chambliss: All right, so my first question is the question I try to ask everybody. How do you define digital humanities?

Nelson:I try not to. I really give very little thought to Digital Humanities as a defined field. It’s like everybody uses that term, so you have to use that term. It strikes me that now it’s more just a… Matt Kirschenbaum says it’s an instrumental term. This is a way of, in a neo-liberal university, being able to brand something that can get resources for humanistic research. I think that kind of makes sense. It’s also like, I’d say it’s like a community, right? DH is kind of the people who call themselves and say, “I’m practicing DH in some way.” It’s just kind of like a way of signaling, “Yeah, I kind of do something similar to what you do, and we have something in common. We’re part of the same kind of community of practice and going to be interested in each other’s work.”

Obviously, it’s humanistic research that involves computation in some way, shape or form, either as a product using mostly the Web or, I guess, apps as a way of publishing and sharing humanistic content and research. Then increasingly, and particularly for Digital History, I’d say, and literary studies, using computation as an aid to doing research, to grapple with big data sets.

I don’t have anything beyond that. I think it can be really overstated. I think it does do some useful work in bringing together people across different disciplinary boundaries, and then that is kind of a sort of interdisciplinary endeavor, but I think that can be overstated. I think some interesting work that’s digital humanities isn’t really humanistic.

It’s humanistic in as much as it’s rooted in some discipline, like history, right? Like Digital History. Sometimes working Digital History is more interesting because it grapples with disciplinary questions than is Digital Humanities because it defies any categorization within a discipline.

Chambliss: I would have described in the intro to this episode that you’re a director of the Digital Scholarship Lab at the University of Richmond. I’m struck by your answer in the sense that some of the things you said are the things that someone who runs the scholarship lab has to say, especially around this question. It’s interesting because my follow up to that question was going to be, and is, what would you say is the value of digital humanities for students, for faculty, and for the public? Because that’s the other thing about this. One of the classic things about digital humanities… I’ve taught an undergrad class in Digital History and I’m teaching a grad class now, and the first question I always ask students to ask themselves, “Does this need to be digital?” Like, this is a really important question. Does this need to be digital? Because all the heartache associated with doing this, if you don’t need it to be digital, just walk away. Right? No one’s going to blame you, no one will ever know. Just, does it need to be digital?  They’re really shocked by that, because they’re like, “We thought you loved digital humanities.” I love a lot of things. That doesn’t mean you need to do it. Right? I like comic books, you don’t need to like them. It’s a question, right? And so the value is really complicated there, so this is really one of those questions that we don’t talk about all the time, but we probably need to talk about a little bit. How would you say that? What’s the value for students, for faculty, for the public, when we talk about digital humanities?

Nelson: Okay, let’s take those in reverse order, because I think that’s easier. The public is an obvious one, right? I mean, like one place that the Digital Humanities and Digital History seems to me like it’s been a success over the last quarter century is sharing research. I guess when this kind of enterprise started, people talked about democratizing free… And in some ways that has happened. That’s a lot of the work that my colleagues and I at the Digital Scholarship Lab do, as I think of the projects we’ve done, particularly the ones on redlining, urban renewal, slave trades, all of them, but particularly I’m pointing out a few that were more successful than others, at least in terms of reaching broad audiences.

And that is one thing that Digital History and Digital Humanities has notable successes at, is broadening, getting people engaged, and sharing humanistic historical resources with not just students. For sure, you get these things used in both K-12 and undergraduate classrooms, but among community organizers and activists and just people who are interested in the history of their own neighborhoods and cities and nation. That seems to me like where, as a form, Digital History has been most successful in reaching that audience. It’s really kind of hard to point at too many projects, because that’s to faculty, right? It’s hard to point at too many digital projects, projects that take the form of something that’s online that has had historiographic significance.

And what I kind of find interesting is it does seem to me there is, I wouldn’t say a categorical split, but the materials that are online and that we would point to as great examples of Digital History tend to be oriented towards the public. The ones that have had an impact for the profession or for the discipline, those use computers as an aid to research, but not necessarily take the form of something that involves a computer. You can read these things on paper. I’m thinking of Cameron Blevins’s article in the JAH a few years ago on Houston and commercialization. What is it called? The geographic imagination. I forgot the title of it. I’m thinking recently of Lincoln Mullen and his colleague whose name is escaping me. They had that piece in the AHR about the genealogy of state constitutions.

Both of those computations, we couldn’t do that research, make those arguments, without computation, or at least you couldn’t remotely, easily, but to get the argument you can pick up a copy in paper of JAH and AHR and read it and the computer was an input, but not a conveyor of that research.

The hardest one you ask, and I think the one you’re most interested in, is students, and I have not figured that one out. This is a mission I have to make. I have never taught intro to DH. I’ve had a couple of DH components in my courses, but I have never figured out how to incorporate these methods in a substantive enough way for my tastes without compromising the humanistic or historical content of the course. If that makes sense, right?

Chambliss: No, that makes perfect sense, yeah.

Nelson: I use some digital stuff and often it’s projects I’ve been involved in, so just last semester I taught a course on the American slave trade and I taught our map on the forced migration enslaved people, but that’s just looking at a map. That’s not teaching DH methods by any means.

And I haven’t… But I’ll back up and say, people who can do that well and can often figure out local projects, getting into archives, building digital collections, building public exhibits, like public Digital History courses, I mean, that’s kind of a no brainer to incorporate digital methods in those courses. The plus, obviously, is that most of our students are going to go on to be professional historians. It’s great that we teach them how to think historically, but incorporating some other kind of harder skills, tech skills, that’s not a bad thing. I don’t think we should pivot and that should be the point of a history degree by any means, but it’s not bad at all to have that.

Chambliss: Yeah, I think my answer to that, I would start with students because when I do digital things, I’m doing it just as you described, in the local context, and we’re doing a digital thing to talk about the local context. Whether it’s going into the archives or trying to piece together the history of a neighborhood and creating some sort of electronic repository around these primary sources, or is it something more mundane but still important, like transcribing the fragments of a long lost newspaper. The digital part there, I always tell students, the tool’s not important. It’s really the thinking that’s employed here in the context. That’s why I always use the terminology critical making. It’s a critical process of making these digital tools, which I borrow from design thinking and some of the work that people do in Victorian studies and making stuff.

But it’s interesting to hear you point to this dichotomy for faculty around the historiographical transformation versus the public and understanding, because it reminds me of how… This is one of the things, at least in my own personal experience, it’s very true. I always point this out to people in my many years working at Rollins College, that it’s not as if the historical questions involved in the digital project are unclear. It’s more that because we had a digital public narrative, people in the community could understand the question. They could see something that, if I just told them or said, “This is the history of reconstruction in the South and Florida’s history is quite bad in this regard,” it’s like, “I don’t believe you.” Somehow you get students to like, “Well, here.” They’re like, “Oh, okay, I guess you must be telling the truth.”

And it’s sort of crossing this boundary between this academic world and the public world. That’s really impactful, but it does run into this problem of why would you do it if you’re an academic? Because your job isn’t really predicated on the public knowing what you’re doing. This is the other thing about this problem.

Nelson: Well, that’s a bigger question about what we should be doing. Because we’re bleeding majors. We’re having less and less impact upon our communities and our society, and being more engaged with the public, that’s intrinsically not a bad thing, it seems to me, and it’s a potential one strategy for increased relevance in a moment where kind of the relevance of all humanities is sometimes not taken as self-evident.

Chambliss: Right, and I think that being engaged with the public is a perfectly reasonable goal. I think one of the things that you hear in conversation around DH is, and you alluded to this sort of neo-liberal, this is a stocky horse to commercialize and marginalize humanities as a critical inquiry into how society operates and the material-driven framework associated with digital humanities, because there are things that you have to make. And I ran into this in a small way in my own institution, only in a sense that I was once asked are you going to keep doing this? And my response was, “Yes, I’m going to keep doing it, and I’m always going to do it about African American community.” Because I was really framing things as a post-colonial DH project where I am trying to recover, explore, explain, document the Black experience in this community, and I’m trying to do that in a way that allows the Black community to have some ownership of it, which was a way for me to justify the work.


But like a lot of people in DH, I didn’t get tenure on anything that was involving DH. It was more traditional stuff, which is a really interesting question and it gets to this other thing I wanted to talk to you about. I think a lot of people, when they hear about DH projects, tend to think of them using computers to do something, which you alluded to in your description, and you in particular through the Scholarship Lab, but you in particular, have done some really noteworthy things with something like topic modeling, where you topic model the Richmond Dispatch. I think that’s actually one of the projects that, when people talk about topic modeling, they go, “Oh, yeah, Rob Nelson, topic model.” Not saying no one else has ever used topic modeling, that’s not true, but it’s really sort of this notable project, in part because I think it was easy to follow and it was easy to understand some of the conclusions that were able to be drawn from topic modeling that particular newspaper.

I think it’s really important to ask you about this. What was your thinking when you hit upon this project? Was this a goal to try to spotlight topic modeling as a technique or was this something that was part of some broader strategy around DH and narratives around Civil War history, because I know that’s a focus for you? How would you contextualize your goals around this seminal project? And seminal is a really loaded word. I don’t mean to say… for those of you who have done topic modeling out there who are listening to this and you’re like, “Oh, my God.” I’m just saying when we talk about topic modeling, your project, Richmond Dispatch project is one that’s often referred to for reasons that I just alluded to. Rob, how would you talk about that project in those contexts?

Nelson: Yeah, that was a well-timed project, is one thing I’d say. I mean, one of the reasons it has been noted by people who talk about topic modeling and its impact upon history and DH is that I got it out pretty early. That came out in 2010. How did I arrive at that? Everything I do is usually serendipity, and that is no exception.

Chambliss: No master plan?

 Nelson: Here’s the three things I could point out. Now they’re eluding me. Three things that came together. One was I’d read Sharon Block’s piece in Common-place. She had a piece out 2007, something about that, where she was using an early version of topic modeling on the Pennsylvania Gazette. It wasn’t LDA, it was LSA, I think. And she was trying, as far as I know the first historian to say, “This has some potential to understand some major phenomena over time and to grapple with big, digitized archives.”

I’d read that and then I looked at her and I think it’s David Newman who is her research partner, and I believe her husband, they had a piece that was cited in there. It said here’s how you do LSA, and the math I realized immediately was going to be over my head. This is not something that I’m going to be able to implement on my own. I probably gave up 10 pages into that.

Here’s the second piece of serendipity. I’d read that article and gotten interested in topic modeling broadly just a little bit before David Bly and David_____ released MALLET. When MALLET came out it was perfect. It’s like this is what I’ve been looking for. This is a command line tool that I can follow the instructions, get all my text formatted, and dump it in. I can do this without having to implement the algorithm on my own or some topic modeling algorithm on my own.

Chambliss: Just for clarification, a command line-

Nelson: Yeah, I mean, I’m still capable of that. I have some technical chops, but I’m not a computer scientist and I’m not a mathematician or a statistician, so some of the really tough algorithmic stuff, not the area I work in. If I set my mind to it, maybe, but maybe not, too.

Chambliss: But even MALLET is, you need to have some proficiency.

Nelson: Oh, yeah.

Chambliss: But you don’t necessarily have to be a theoretician. So that’s what I mean. When you say command line, what does that mean? It means that it’s like looking at DOS Shell, if you remember DOS Shell, right?

Nelson: And what it gives you, I mean, this is the difference, too-

Chambliss: No one remembers. That’s a stupid example. All right.

Nelson: I mean, it will give you a ton of just text files out of it that have a bunch of numbers in them and you have to be able to do something and put those numbers… I put them back in to transform them into a relational database that I can put online and then I can build an app on top of that to visualize the topics that are discovered there. I don’t want to, no, not everybody could just use MALLET. You need some knowhow even to use the outputs of MALLET, or even to just get MALLET to work in the first place. But the third piece of serendipity is that I’m a 19th Century historian who found myself at the University of Richmond, and a few years earlier my colleagues in the library had digitized the Richmond Daily Dispatch. They’d gotten an IMLS grant to digitalize the whole run of this thing. Within the library, I could walk to one of my library colleagues and say, “Can I have all those text files?” And he would give them to me. He would give them to anybody, but it was really easy for me. I knew about it. It was just kind of perfect.

Those three things together. The Digital Scholarship Lab, we’re a little more strategic. We have a little bit more of a plan. As director, I wouldn’t say we have any master plan or anything like that. That’s a loaded term to use. But in 2008, ’09, ’10, when I was doing that, the emphasis was more on the lab part, and by that, I mean experimenting. We continue to try and do interesting things. I think we’ve got a little bit more of… we’re a little less just kind of screwing around at this point and we are probably much better at figuring out what kind of projects we want to work out and having them lined up and knowing where we’re going with them.

I did not know where I was going with mining the Dispatch, partly because the method was so new. Which is why that project had as much impact as it did. This was an easy thing for people to see that you couldn’t see just with the output of MALLET. For that project, I always try and just build things that I want to use and I wanted to understand the method myself and I wanted to understand the Dispatch better. I wanted to understand Civil War Richmond better, and so I kind wanted to see what I can do with this topic modeling, and mining the Dispatch was the first product of that.

Chambliss: Right, and as I say, I think it’s still impactful, but your answer sort of leads me to my next question. I think much of the digital humanities work that’s associated with the Digital Scholarship Lab I would describe as digital recovery around US history. I say that because I think in particular a project like American Panorama as a project that has lots of little mini things underneath that umbrella. That’s sort of like a project umbrella. When you look at the project, I’m always just struck by they’re providing a kind of context to American history that people know, clarification around the history that people know, and it’s part of the reason why they’re so impactful, I would argue, because there’s already a kind of baseline public knowledge around some of the things that the projects are focused on in American Panorama, but you provide a kind of illumination. You’re recovering what people might recover in a sense that academics might know some of these arguments, but you’re really visualizing it in a way that’s really compelling.

Because I think that project is so important… And that’s also one of the projects that I know that you guys get a lot of attention for, because every iteration I’m teaching a class now and I’m using Slate’s list of best digital projects that they’ve been doing and you’re always on it. I know you say you have no plan, but I feel like when you look at American Panorama, there’s a plan here.

Nelson: Yeah, that’s where we’ve gotten a little more intentional about the work that we do. History of American Panorama. American Panorama, that was Ed Ayers’s idea. I don’t know how long, but long before he even came to the University of Richmond, I wanted to work on a project and help propel a project that would use the web and use interactivity of digital media or a spatial mapping project on US history. I will take basically minimal or no credit, because I don’t really deserve any. It’s all Ed. A few years, six, seven years ago, got a $750,000 grand from Mellon Foundation to help us start that project.

We worked with some great tech people out at Stamen Design in San Francisco, which was illuminating for me. It was like an education. I mean, I came out of that process much more capable of, or at least on a trajectory of being capable of doing really complex web applications, which we’ve done. They built the first four maps of American Panorama, which was our forced migration map, foreign born, one on the overland trails, and another one which people tend to ignore, which is canals. Don’t ask me why they ignore canals, but [crosstalk 00:25:19] doesn’t get a lot of traffic.

Since then, we’ve developed in house four other maps and really, in some ways, because of the topics, not because of the technical development, we have our map on redlining, a map that by far dwarfs everything else in terms of the amount of traffic and probably impact that any of these have had. We had a follow up to that on urban renewal. That was released a year and a half ago. Recently released one on electing the House of Representatives, and then a smaller one, which I think is kind of a fun little one, which is on travel abroad of secretaries of state. It’s called executive abroad, so it’s kind of America in the world, at least as its expressed through executive travel.

It’s a massive project. There’s a kind of preface to this, a prequel to it, which was our enhanced edition of Charles Paullin and John K. Wright’s Atlas of the Historical Geography of the United States. That was us doing our due diligence by looking at what had been done with the genre of the historical atlas to date. That was kind of the print model for us. We wanted something of similar ambition using digital media that Paulin and Wright had done with their 1932 atlas. It’s proven to be more of an analog that maybe we wanted, because that took them, depending on how you date the work on that project, two or three decades for them to finish that thing. It was kind of a nightmarish project that barely got finished. We’ve got a long way to go, too. We’re seven years into this thing.

I think saying they’re maps is a big underselling them. Each of them is thousands, hundreds, thousands of maps. But, still, there’s an unending number of topics in American history that we could and should tackle spatially to really make this atlas remotely comprehensive. Nothing is comprehensive, but we don’t have the presidents. We don’t have presidential elections in there yet. We have a very early project on that, but we can’t have a finished atlas that has Congress and not the presidents. We can’t have that. We’ve got to get churches in there. We’re going to have to get agriculture in there somewhere.

Chambliss: Yeah, it’s a great project because it can go on and on and on.

Nelson: It could be never ending.

Chambliss: Yeah, right, exactly, and it’s interesting because whenever I look at that project, again, when we talk about Digital Humanities, my experience is that, one, no one knows what it is. This is a thing, no one knows what it is. Going back to your original answer, that term is really broad. But once they decide that you do it, then you do it, and then they start asking you stuff. I’ve had a couple of conversations in my own institution where I was like, “Do you know how much money?” Or could it be like, “Do you know how much money? They have millions.” I work alone with students. No, it can’t be like that, because they need resourced to get a certain feel and look in order for them to hit this mark. This idea that an individual doing digital humanities versus a team doing digital humanities, which lab is a really loaded term in DH, and the Digital Scholarship Lap at UR is one of some known labs. It’s a known entity.

Which gets at this other question that as, again, someone who is running a lab funding, more importantly the search for funding, is a very known thing. It’s like an inside baseball thing for people who are doing the age. It’s a problem, it’s a challenge, it’s an opportunity, depending on how you want to look at it, depending on the day, how you got out of bed. It’s one of those things. As director of the Digital Scholarship Lab, you have captured, I would argue, some significant grant funding. How did that challenge vex you? Is it haunting you? How does that work?

I mean, yeah, you apply for a grant, but one of the things about grant funding that probably the public doesn’t know, once you get a grant, it’s easier to get a grant. When you don’t have a grant, it’s way harder to get a grant. American Panorama is a great project, because it can go on forever.

Nelson: Yeah, but we haven’t got anymore money for it.

Chambliss: What?

Nelson: And I don’t think that’s necessarily a bad thing. We haven’t really searched for more money for it. I’m not discounting that money. That seed money was unbelievably important for us, right? And we could not work with Stamen. They were, frankly, expensive. I’ll just leave it at that. It gets very expensive when you’re working with somebody in San Francisco or Silicon Valley, and it was worth it, because they upped our game significantly, but since then everything is built in house. We work on next to no, my discretionary budget is next to nothing. We have a couple software things. That’s probably the biggest thing is I spend a chunk of change every year on Carta DB so we can use that to simplify some of our workflows and don’t have to host certain things here. I have been of mostly one mind. Let’s say one and a half minds, because it’s not quite two minds, about funding. Obviously one of the big funding agencies for people doing DH is doing it through NEH and ODH, the Office of Digital Humanities at the NEH. It has been a while since we applied, been a long while since we applied to anything from them. The reason is, like, A, I am in a really enviable position where I have a small staff and nobody’s on soft money. I don’t have to fundraise to my staff, which is great. It’s like, oh, yeah, my God, I mean, the stress people are under when you actually have people whose paychecks are dependent upon you getting money. I just, that gives me cramps just thinking about being in that. That responsibility is horrible. I don’t have that.

Chambliss: And I think it’s worth clarifying for people, a lot of positions in all research centers at academic intuitions are soft money, which means that they’re grant funded. When the grants run out, the position goes away.

Nelson: Exactly.

Chambliss: And it’s usually a small percentage of people who are hard money, which are faculty or librarians or staff people who are part of the institution’s budget and they really rely on the ability of the director of the center to raise money, to write grants, to raise money, and almost all these truly massive centers… And I’m actually kind of shocked that you guys are all hard money, actually.

Nelson: Well, the DSL is less than three people, because it’s me. I have a GIS analyst and project manager, Justin Madrin, and I have a visualization designer, Nate Ayers. And the reason we’re less than three is that I’m now head of digital engagement in our library, I oversee digital collections and digital preservation. It’s not like my time is split 50/50 between the digital engagement and DSL. I still lean towards the DSL because I do a lot of that, a lot more involved in the nitty gritty of that work.

Chambliss: So you’re not a center?

Nelson: But it’s not a big staff, right? We’re not CHNM with dozens of people. We’re not even MITH, right? MITH’s got probably three or four times our staff, as it should. It’s a big research institution. MATRIX, I don’t know how many staff there are at MATRIX, but I’m guessing it’s bigger than the DSL.

We’re modest, that’s great. Coming back, one reason I’ve been hesitant about funding is, first, I don’t have the stick of having to fire people. Having to fire people, that would light a fire under me to go get grant money. But the amount you get from NEH, by the time you write the grants and do all that planning, you’re ready to go just get the project done and you turn it in and you wait months and months and months and months for them to evaluate it. Then you get it and there’s a couple months wait before you start getting money, right? For us, I just like doing the stuff, not writing a grant so we can wait to do something. For us, it just never has made that much sense, because I’m not dependent on it. Administering grants is no small amount of time. Writing them and administering them for me almost mitigates the actual benefit of getting that money. Where one of the benefits comes in is, well, this comes back to the neo-liberal institution, too, or one way of saying you’re bringing money to your institution. Your institution might care about that. Another way might be saying in a profession that’s been defined by peer review and you’re working in an area that doesn’t have a clear substitute for that peer review process, grant funding can serve as a surrogate or a proxy for that.

You can point and say this is me being peer reviewed. This is me submitting my work to colleagues and other practitioners in this enterprise and they’re saying this is good stuff and we’re going to support it. And I don’t have that, because I’m not on a tenure track. I wouldn’t mind those accolades, but I don’t need them. People who are waiting to come up for tenure or trying to get to full professor, that might help them to have that and be able to list that in their tenure promotion documents. For me, it doesn’t matter.

One thing that I like about the DSL, and I feel it’s not for everybody. We’re kind of idiosyncratic. I never want to suggest we’ve figured this stuff out. I would never suggest that. At the same time, for us, getting grants has not been the only or even the, is pretty far down our list of priorities. Getting work done, that’s our priority. I don’t know if that makes sense.

I’ve got to come back to one more thing you said, because I think you raise an interesting point about groups versus individuals. Seems to me like some of the most interesting work that’s happening in Digital History and the Digital Humanities is by younger scholars who are probably a half generation younger than you and me who know how to do this stuff and they don’t rely on anybody. They just do it, right?

Chambliss: Right, right.

Nelson: They own their own stuff. Output some stuff. They got PhD’s in whatever, history or whatever field. They interpret the results and they write things, and it’s not like they need a team to do this. They just do it.

Chambliss: And that is, in fact, the norm. That’s true. When you think about what the future  graduate from a PhD program will be able to do digitally, they’re going to be able to do a lot. They learn the tool early in their career and when they’re doing their research, they’re able to look at the data they’ve collected and they can tell us with that data, with the tools that they know. That’s exactly what you see younger scholars doing. They’re telling these tales and they have a dissertation that’s associated with it, or they have the digital dissertation, which is incredibly something that is more and more common, which has its own set of complications, but it is a thing.

So, yeah, like at that place on MSU, that is one of the things that you see happening. You see a large number of students coming in, articulating in their documents, “Oh, I want to do this thing. I’ve done stuff as an undergraduate, I want to do this, I want to do that.” And that was one of the things I would tell our undergrads at Rollins, you’re going to need to know one tool. Just pick it, pick a tool. Doesn’t matter what the tool is. I was super liberal about this. Tools are the things that you use to tell a story. You use a pencil, that’s a tool, so just pick a tool and if you want your tool to be, I’m going to make podcasts, all right, then you need to learn editing software. Also, it’s really about you need to be able to sustain this. You, yourself. You need to be able to carry this water all by yourself. There’ll be no one there for you. You have to be able to sustain this.

Nelson: That’s good advice.

Chambliss: If you can’t do that, then don’t do it. And they were always like, “You’re being really pessimistic.” I’m like, “No, I’m not, because everything costs money and if you do this thing with elaborate software that we have in our labs here with a 3D printer that we have here, you’re not going to be carrying a 3D printer with you everywhere you go.” Just the nature of the beast. That is the reality. But we’re at a pivotal moment, because you have a lot of scholars, as you point out, who can do these things and we’re only just now as a profession going, “Yeah, this is how we’re going to value it. This is how we’re going to evaluate it. This is how this count for tenure.”

Can you do something digital for tenure is a huge question at a lot of institutions. They’re making the transition. We’re riding those things right now. For the most part, we’re moving in the right direction, but the cost problem, that is a question, because I think increasingly the real question here is the thing that you do that’s digital has to be truly original in order for it to count. It has to be this thing that is deeply grounded in your research, because the tools are there for almost anyone to make a really interesting visualization, and they may not know anything about visualization or the logic of digital [inaudible 00:41:24] or just hind thinking. They can just put stuff in and it’ll produce this spectacular looking thing. It’s really meaningful in order to distinguish yourself and that baseline becomes really complicated. I don’t want to keep you too long, but my last couple of questions, the reason that I’m able to talk to the great Rob Nelson is because of the History Engine project, which is a project that I started working on more than a decade ago. Yeah, more than a decade ago.

Nelson: Yeah, I think it was 2019. It was 2008.

Chambliss: Right, and this last semester at MSU I had students do entries in the History Engine. What’s the status of the History Engine as a project?

Nelson: Okay, I want your opinion on this, too. This is actually perfect, because I was having a conversation. I used it in my class this last semester. I really like it. My students seemed to like it. They got more out of it than writing conventional four-page response papers. They seemed to really enjoy it and writing for a public and doing that kind of storytelling, I think they got more out of that little modest research assignment than they did out of something super conventional and a typical response essay, though I had them do that, too. But they seemed to like the History Engine better and seemed to get more out of that.

But I have been thinking about writing to everybody this semester and saying we’re done. It won’t go away, but we’re not adding to it. Let me tell you what my thinking process on this. What kind of prompted this is somebody wrote me and said this isn’t working and it’s because we were using Google maps to show the locations where these historical vignettes took place. It had escaped me that they moved to a kind of really expensive… the free version basically went away, right? Had to sign up and use a new library.

The library had been discontinued, so I had to recode it and then I potentially had to pay for it or find a different solution. It’s like, “Crap.” I got new stuff I’m doing that I don’t really want to be working on the History Engine anymore and I’ve kind of forgotten how some of it worked. Any time you’re revisiting a project that that’s old. Kind of the amazing thing about it is it actually hasn’t collapsed. Ten-year-old project that really we’ve done minimal maintenance on it and it just kind of seems to work, but I feel like that can’t go on forever. My main reason I’m at least considering shutting it down for future contributions is I don’t want you or anybody else to have this built into your class and then it breaks and we’re struggling to get it, or maybe unable to get it working for anybody.

If you’re going to offer this and say that people can use this in their classes, it’s got to work. And I feel like we’ve dodged a bullet to date that it hasn’t broken at some inopportune moment. And the other thing I guess I’d say about the History Engine is that it certainly gets a lot of traffic and it’s gotten very modest but steady use. You can tell people are using it every semester, but I think when we were designing that, we thought there would be some kind of geometric growth in this, and there have not been. Five, six, seven people use it every semester, and it never is more than five, six, seven people. If it had a forward momentum, I would hesitate about shutting it down, but it doesn’t seem to. It’s good, it’s fine, nothing wrong with it, but it doesn’t have a great deal of growth in it.

What do you think? I’m not talking about taking it down, I’m just talking about shutting it off so you can’t log in and add stuff. What’s there would remain there and be kind of a simpler project, but it wouldn’t be a growing, living project. Would that break your heart?

Chambliss: First let me say, History Engine is a project where students in classes across the country work in an archive to write what were defined by Ed Ayers as episodes. So like narrative vignettes of history drawn from local archives. Many institutions, many organizations have archives, but most of us don’t know what’s in them. This is a way for students to go into archives, be historians, work with them, write these sort of contextual narratives that shed light on the past.

I’ve always enjoyed doing the project. A lot of my own thinking about Digital Humanities sort of spun out of it. There’s something in the History Engine that I would be like, “Oh, I can do this. Oh, these fragments actually represent this,” and built projects based on having worked with the History Engine. Yeah, there’s a part of me that’s like, “Oh, no, I don’t want it to go away,” but, yeah, I totally get what you’re saying. When we started doing it, I think we started doing it 2007. It was 2007. I think that was the year, because this is how old it was. We got a grant from an organization that no longer exists.

Nelson: Yeah, NITLE, yeah.

Chambliss: Right, yeah. National Institute for Technology in Liberal Arts Education. Like, they’re gone.

Nelson: Yep.

Chambliss: They died. From a digital standpoint, the History Engine is ancient. Right? It was updated, I would say significantly updated five years after that 2007, maybe around 2011, 2012.

Nelson: Something like that, yeah. We did a redesign and just kind of modernized it a little bit.

Chambliss: You guys put out a French version.

Nelson: We gave the code away to anybody, and so somebody in France wanted it and we gave it away to them.

Chambliss: But, yeah, it’s no longer the new hotness that it was, right? It just isn’t. Yeah, keeping it alive isn’t a thing. I used it this semester because it’s a great project to have an outcome that’s rooted in archives. Like even now, I’m at an institution with the world’s largest open collection of comic books. That means you have the largest publicly accessible collection of comic books in the world. So this semester we went in. Everybody wrote articles, well, episodes based on a corpus that I put together for one of the sub-selections all about superheroes. And it was a great project, because it got them to work with this material that we were going to work on anyway for another project involving digitization of those same objects.

I needed them to get familiar, I needed them to dig into that, I needed them to think about historical context, I needed them to use primary and secondary sources to make sense of this material. The History Engine is a really great way to do that. It’s a really great way to make a person stop and go like, “This object, this primary source, opens the door to numerous ways to talk about the past and I’m going to talk and I’m going to talk about it this way. This is why.” It was super helpful, because when they came back later I’m like, “You know what you did in the History Engine? Do that, except add this stuff.”

Because they had the right descriptions of objects that they were scanning and I wanted them to be historically contextual, like rich descriptions. Not just simply descriptions you would find in most databases. It’s a book. The cover describes this first. In that moment in history, these things were happening related to this character around comic books and blah, blah, blah, this anniversary, blah, blah, blah, so that when someone saw it in digitizing and look at it and go, “Okay, it really connects to this moment in comic book history.”

They wouldn’t have been able to do that except that whole process of using History Engine to prime them to think about it in a certain way. To me, it’s a great tool, but, yeah, if you got to keep it alive, it’s just a thing.

Nelson: Yeah, it feels like a time bomb sometimes. It’s going to go at some point and I got to defuse it before it, right?

Chambliss: Right, before it all goes to hell. The engine’s going to blow, captain.

Nelson: Exactly. That’s perfect. Yeah, I got to play with that metaphor.

Chambliss: I understand the problem, I do. I wouldn’t want it to go away. I mean, one of the things about history over the years is that people will contact me about an episode that was written in the class. Students tend to enjoy writing them.

Nelson: Yeah, that’s been my experience, too.

Chambliss: Because I’m often like, it’s not a fictional story, right? The students love the idea of writing fiction for some reason, even if they’re not English majors, but especially if they’re English majors. I’m in the English department now. It’s like, “No, it’s not fiction. It’s a narrative. It’s a contextual narrative that’s grounded in things. Don’t make stuff up.” I understand the problem, and in some ways it’s a legacy project and this is one of the things that we all have to think about when we talk about digital humanities. What do you do when the technology moves on and you still want to try to maintain the project? How do you move from the active collection of data and actively doing it to a kind of legacy status? What does that entail? How does that work?

When I left my old job, I had a ton of stuff online. They were like, “This is all going to have to come down.” Because I’m not there anymore. What do you want us to do with this? Do you want the files? I’m like, what do you mean it’s all got to come down? It’s a perfectly reasonable response on their part, they’re perfectly nice people. We’re going to crawl it and then it’s going to come down. It’s going to go into [inaudible 00:52:32] and it’s going to be like, but it’s going to look ugly in there, I said. That is real. That is real. I understand. You should probably stop, but-

Nelson: Okay, good, good. That’s the answer I wanted to hear.

Chambliss: Yeah, you should stop letting people put stuff in there.

Nelson: You’ll get in an email, the last hurrah. Do it for the fall.

Chambliss: I think you should have a party for all of us and invite us up and we’ll drink some champagne.

Nelson: That’s what I should get grant funding for, little wake for the History Engine.

Chambliss: No, it’s not a wake. It’s a remembrance. When I go to the DSL website I want to see a little in remembrance sticker for the History Engine, and maybe some Civil War music playing.

Nelson: It’s going the way of NITLE.

Chambliss: Because it lived a good life and it fought a good fight.

Nelson: It’s had it’s day.

Chambliss: No, yeah, I think you can-

Nelson: Do you ever get, you mentioned you get contact, and this is by students or do you get contacted by people who see? Because I get the complaints. Like this semester I got, it was super interesting exchange I had, and this is maybe unsatisfying for the person I was corresponding with. Somebody wrote to complain because there was an episode that was from probably Ed Ayers’s class a long time ago referred to the bravery of Andrew Jackson as he was fighting Native Americans. A Native American woman took a great deal of offense at this, understandably, and she said this is kind of celebrating him. I could have nitpicked and said you can be brave in a bad cause, but I didn’t want to be nitpicky and said I’m with you on thinking Andrew Jackson did a lot of evil things.

Chambliss: Not brave things, yeah.

Nelson: Yeah, the displacement of Native peoples is a product of greed and racism that we should be nationally shameful, but I’m not going to take it down because it’s written by students and unless something is, I wouldn’t let hate speech stay up there, and I will take things down if they’re just clearly-

Chambliss: Wrong.

Nelson: Like I want to give them the story. It can’t be fake history, it has to be a reasonable, not completely erroneous version. Students make mistakes and I let some of those mistakes go, but if something’s just totally off I’ll take it down. But in this case it wasn’t off. I wouldn’t have said that necessarily, but I’m the editor. I’m not going to have that kind of heavy handed editorial control over the project. At the same time, I’m kind of with that woman in a lot of ways. I don’t want to keep writing these emails. This is the other thing about the History Engine, is I have to defend this editorial policy, sometimes when I find it troubling to do so. Let’s take it down so that no other student says something that’s going to offend somebody and then I have to defend it.

Chambliss: Well, you know, that particular case, I don’t usually get complaints. I usually get people asking to follow up on something. Students typically don’t necessarily contact me about it. It’s interesting to third parties. I don’t get that, but a lot of these are private. When you put stuff on the web and you’re working with students as collaborators, you get a lot of complaints. I’ve gotten my fair share of complaints. I know, I get your interpretation. Is that interpretation wrong? Sometimes I just take it down because it’s an investigative project. A lot of my projects are locally shared investigative projects and I don’t need to argue with this person. Other times I’m like, that’s an interpretation that’s a little eh, but it’s okay. With History Engine episodes, I will just unpublish them.

Nelson: Not everybody is as responsible about it or cares as much about the History Engine. That kind of investment in this, and that’s always been a kind of tough thing, too. How do I put this? Because I don’t want to be elitist, right? Because you at Rollins, Catherine at Wheaton, Lloyd at Furman.

We had good students doing this work and not everybody who’s written has been a school that has as strong students and I still, because it’s pedagogically oriented, I always make for any college, doesn’t matter what college it is, an account. But it means we do have slightly varying degrees of quality in there and sometimes people, I don’t think, they’re not going to unpublish. They’re just going to let their students publish it and if it’s junk, it’s-

Chambliss: Right, yeah, right. It’s complicated, because you want it to be an A paper. You want it to be an A paper that’s in the History Engine and then you go maybe it can be a B plus, maybe it can be a B. Then after that, I’m like no.

Nelson: That’s exactly my cut off. B or better it stays there.

Chambliss: Because then it’s like adding injury to something that’s already injured. I’ve kept you a long time. My last question for you is what’s coming up for you in terms of Digital Humanities? What’s on the horizon?

Nelson: It’s actually kind of retrospective, I’d say. At the moment what we’re working on, mapping inequality has been the most visited project that we’ve had. I’m an urban historian. You’re more of an urban historian than I am, but it’s had a big impact. People like. It’s a window into wealth and racial inequality in their communities. It gets widely used. It’s used in schools, it gets used by activists. The research that’s come out of this thing has been kind of mind blowing, even on the spatial data.

So the next thing we’re working on is we’re revisiting NAS. We’re going to add three dozen draft maps that aren’t up there. Then the big thing, which kind of is a new direction, is we’ve gotten really close to having all the area descriptions for that transcribed. It kind of comes back to the beginning of our conversation, I’m interested to do some text mining on these. I’ve done a little bit of preliminary work using these, but we’ll have about 8,000 of these documents, each of them has about 50 data points. Not every one of them has that, but there’s a lot of data there, and some really interesting things that we haven’t explored that we’ll use that textual material connected with this program. I want to make it possible for other people to use that, too.

So we’ll obviously share the data eventually, and we’re going to have search. Throw in a word. You can throw in smell. I’ve gotten interested in smell, because a window into environmental inequalities, too, but working-class people. They were near flower houses, factories, it stunk, and these get noticed. And so you can kind of get a sense of the environmental distinctions and inequalities within a city using these. We’ll do that.

Then I don’t know exactly where it’s going to go, but I want to surface larger patterns and see if we can… We, and I mean that like us, and I mean that as the broader community when we share the data, can do something as important or as impactful of the area descriptions as we’ve been able to do with the spatial data to date.

The other thing I want to do, which we haven’t, it’s not disciplinary and it’s about the medium and not the topic or the material itself. None of our maps work on mobile devices at all, phones, and this is when I want it to work. That’s been okay, because we want a big canvas and a lot of these are so interactive and so much data that it just doesn’t really lend itself to a little tiny screen, and I’m kind of okay with that. It doesn’t have to work on everything. This one, I want to work on everything, because I want you to be out in a community and be able to look at the area description for the area around a downtown that you’re in and see the horrible things-

Chambliss: Thanks a lot for taking the time to talk with me.

Nelson: Thanks, it was fun.



Icon for the Creative Commons Attribution-NonCommercial 4.0 International License

Reframing Digital Humanities: Conversations with Digital Humanists Copyright © 2021 by Julian Chambliss is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.

Share This Book