 And he is widely published. He's the author or editor of four books on America in the mid-19th century, including House of Abraham, Lincoln and the Tads, A Family Divided by War, the Book of the Month Club Main Selection for March 2008, and Weirding the War Stories from the Civil War's Ragged Edges. He oversees the web project CSI Dixie, devoted to the coroner's office in the 19th century south. He is the secretary-treasurer of the Southern Historical Association and the co-director of the Center for Virtual History, and co-editor with Amy Marie Taylor of The Uncivil War Series at University of Georgia Press. So without any further ado, Rappessa Stephen Berry. Thank you all. Tech work. So I'm an historian, obviously. I'm going to try to make things relatable to where you are in your process, as you embark on your DH projects, but you're going to have to help me along the way. So I don't talk that much about historical content. At the same time, I'm an historian, so I'm obsessed with the content. And so I'll do my best to make it relatable. But really, I want to talk about process, and I want to talk about what didn't work, especially, because I think that's where we can be helpful. We failed a lot. So we founded the Center for Virtual History in 2010 with Claudio Sont, who's an historian here as well. He works on Native American studies. And for at least the first four years, I called us not the Center for Virtual History, but a virtual Center for History, because we virtually didn't exist. And we just flailed around. I'll be very honest, and had a number of false starts. Our name, the Center for Virtual History, what we thought we were going to do is really virtual history, like virtual reality. So we thought we would take 19th century photographs and use Google SketchUp to drape them over buildings so that you could walk through a 19th century to get a type city, essentially. We're going to do Pose, New York. We're going to do Lincoln's, Washington, DC. And we're going to do Atlanta in 1864 before it's fall. This is a great idea, and this is what I'm telling you. You start with a great idea, sometimes it ends there. And that's what happened. We just didn't have the time to make this work. We gathered as many photographs as we could. We put as much time and effort and money into this. But we need cave technology, right? To do virtual reality, we'd need to be a Georgia tech. We just didn't have it. So we failed, like right out of the gate. And our name still bears a mark of that failure. So then we regrouped. And our first project that we released in 2012 was Indian Nation, another great idea. So I'm obsessed with the census, like nothing can get me to geek out more than the census. 1900 is the low point of the Native American population in the United States. So from 8 million people, right? When Columbus took his toes in the Bahamas, we're at 237,000 Native Americans captured in the 1900 census. And the 1900 census also has a separate questionnaire that goes to Native American families and asks them additional questions. 1900, right? We're very close to the Dawes Act. The Dawes Act is essentially gonna dissolve Indianness in America, dissolve the tribal council, send the kids off to school, cut their hair and produce as they thought Americans. So you can just get an example. This is Tom Torlino, but they take this picture. This is a publicity stunt for whiteness, essentially, right? So they've taken him and they've made him not just a model Indian, but they overproduced the photograph to produce a whiter face. So this is what's happening to America in 1900. And yet, somehow, they hold on to end your mess, right? And now there are 5 million Native Americans in the United States. So from this low point in 1900, we were almost gone today. And it's because of this generation, right? That held on to their communities and their identities. And so our idea was to do crowdsourcing. So there were all these citizen science in 2012, there was a big movement, citizen science, get your amateur astronomers out in their, outside of their garage to help map the heavens. And it was working, it was really working. And so we decided that citizen history could be a thing. We could do this with crowdsourcing. We got an agreement to work with the Cherokee in Oklahoma to set up kiosks to take in data for all of these 230,000, 7,000 people and data about their families. So the way it works, if you just search people, these are all the graces in the database. We'll just click on Ada Grayson. And there is no data. It's crowdsourcing project, right? So all we did was sort of a rip and run on a census. We pulled the data that we could get out of the census. But then we were counting on the crowd to help us fill it in. We also thought that this isn't a database of individuals, it's a database of communities and that was really important to us. So we also have everybody essentially in communities. So this is obviously a large one in Custer, Montana but you get these tiny young clays. One member in Aiken, South Carolina, there's one Native American trying to hold out as a Native American, one of 237,000. So we wanted to tell this story of this generation that went through this inflection point. It's a great idea, it didn't work. And this is a lesson, right? If you build it, they will come model is not a thing. It's just not a thing. If you build it with them, that's an option or just build it yourself all the way through. But trying to do crowdsourcing after the fact was a disaster. It might have caught fire if we'd gotten, we'd put in for a couple of rounds of NEH money. I don't know how they could understand, to be honest with you, but they did. And so this is going to be warehoused, probably. We'll probably keep the, there's the other thing about this early days, right? So this marries a Drupal CMS. Did you guys talk about Drupal and content management systems? Okay, so a content management system is exactly what it sounds like and it's not as fancy as it sounds like. WordPress is a CMS. It's just a way to have a back end to enter things in that then is cast at the public. I'm gonna talk a lot about Drupal today and go inside Drupal because it is my favorite CMS and even humanity's types can use it. Just as a quick thing, Drupal just is basically a PHP writer. It's just a sequence of menus and modules that allow you to write PHP so that you don't have to code. PHP is the language of the internet, right? And so I find it. I don't have to teach myself one arcane of language. Python would be awesome. If I knew it, I don't know it. Ruby on Rails would be awesome. I don't know it. I barely know HTML, actually I do, and I barely know PHP pretty well, but I don't have to write it if I'm using Drupal. A White House.gov is on Drupal. The libraries here at UGA are on Drupal. Our history departments, web page, all the web pages of every department including containers, all on Drupal. So it's a huge open source CMS. This thing marries that to ArcGIS. So we've got a window in here that's casting ArcGIS for our communities and all of the little levels over the communities and the link to describe the community, expensive. Very expensive, because it's not out of the box and it's in Frankenstein. It's cobbled together a Drupal CMS on top of ArcGIS and trying to do the pass-throughs of all the data. It probably cost us $25,000. Projects, I can now spin up a project. If the data's good and somebody gives me a clean CSV or Excel, I can spin up a project in an hour for nothing. And that's where we are with the technology. There really isn't this barrier to entry that you used to think, oh, I've got to apply for a bunch of grant money to get this done. No, you don't. You teach yourself a little bit, not even to code, and then you can get something working. Now, it might not be great and we all know there are a lot of derelicts and disasters that are out there on the internet that are not quite professional and wouldn't pass peer review. We can talk about that process too as we go on. Nevertheless, it's gotten easier and easier. Okay, so this will probably end up being shelved over at ArcGIS Online, AGL. So we'll park the mapping portion of where these people were, their communities. But whatever data was in here about the individuals will probably be lost unless somebody wants to take this project over. Okay, so we learned something, a lot of what not to do. And this was our third project. So we'd failed twice, I mean, conceptually, like at the level of our identity. So we're the Center for Virtual History, we're the Virtual Center for History who's gonna do virtual reality and we have no tech. Bad idea, right out of the gate. Second idea, we're gonna do crowdsourcing without really, I mean, we're just little monks. We don't know anybody. We don't even know 10 people, much less than 100,000 people who are just not popular enough. I don't know, I don't know what happened. But so conceptually, twice out of the gate. We're not a virtual reality company or entity. We're not a crowdsourcing entity, they didn't work. So our next project was Pax Americana, which maps the 1775 to 1782 smallpox epidemic in the United States. It's drawn from the data of Lil Fen who was, I think she was then at Duke, she's now in Colorado. And she wrote this book called Pax Americana, the smallpox epidemic of 1775, 1872. Her argument was that Cornwallis and George Washington are not playing a guess game of chess against each other. They're both playing a game of chess against the third party, smallpox. Whoever manages smallpox better wins the American Revolution and the American Revolution's inflection point or turning point is much more about disease than it ever is about pinning Cornwallis in Virginia. Anyway, so this is kind of a weird afterlife for her book, right? She publishes her book to great claim, but she has this data set that can live in the world in a different way that we can take apart. And this is what I like about DH is really thinking about structure and form, of how that knowledge wants to live in the world, how the argument wants to live in the world. Form follows function. So in this case, you can watch this as an animation so that you're sort of watching the epidemic unfold. Not really, right? Because what you're watching are not cases of smallpox like we know what you're watching are the data collecting points, mostly that write in English. So is smallpox happening in the middle of America in a particular tribe? Sure, is there one white dude there to write it down? No, so we don't know about it. So we're mapping the data and the data collectors, but we're not actually mapping the epidemic. So it's got a conceptual flaw sort of that, it's hard. But it is interesting because what it does is it takes her book and deconstructs it essentially. So I call these kinds of projects a deconstructed monograph. You take a monograph partly apart, right? So a monograph is an argument, a monograph is a collection of data, a monograph is an argument about the data, and you break it apart enough that it can exist in some new way, in some new combination and permutation that the user gets to impose a logic to or tell a different story or a different narrative arc or a different argument with her data. You can do that because it's been taken apart, but not fully apart, we're not back at the archive. All of the work she did to collect thousands and thousands of points about smallpox in America is right here. So this is a probable case, we're not sure it was smallpox, we know it was disease, we're not sure quite that month and day, it's roughly this, and then you get a quotation and a citation and that's all of what this is. So if you wanted to tell the smallpox epidemic from the perspective of African-British, right? Who actually are devastated by the smallpox and have been in the British actually just quarantining them on an island of disaster mayhem and arduousness, but she doesn't really tell that story, but you could decide you wanted to tell that story, or you could say, well, I actually do want to tell the story of George Washington. So you can go and search the individual reports looking through the citations simply for Washington. They should, I'll come up in a second, our connection as long as it was working. Anyway, I don't know what to say about this project in terms of like the lessons that we learned, it works, I like it, nobody uses it, it's nobody's baby, right? I think this is one lesson, you need one person about this project to be obsessed with it, with adding to it, like you need an OCD component to, especially on the data side, you don't even know what's driving you to collect every one of these. It's weird and it doesn't need analysis, you just harness it essentially. And since this was somebody else's day to begin with and we spun it up because we could, then it didn't have an owner, it didn't have a parent really, it didn't have a public, it didn't even have a scholar. The scholarly arguments have been partly made. I think people could play with it in a classroom setting, I think that could be its best use. So thinking about environment and disease and war and how we write military history or would better write military history even of this founding event of American history. But it didn't work, not really. Like we got neither credit for it nor as much satisfaction out of it as I'd hoped. And this is where Claudio and I started to partway is not in a negative way, I think in a positive way where we were both figuring out who we wanted to be, like how we wanted to be digital scholars. And we hadn't figured that out before pretty clearly. He figured out that he wants to act stuff. And I kind of do, but not really. I have a different interest in digital history, a different sense. It's you're exploring forms and formats, right? Really for the first time. At least for my line of work, it's a, you have an idea? Oh, is it a book or an article? These are your forms. There are no other forms. It's an embarrassment considering how creative you are about all the work that goes into collecting the data and making the argument on the rest to not even consider what the form or format and how the thing wants to live in the world and to really be starved down to two options. It drives me crazy. And so what happened, I think, is that we both sort of had an awakening into becoming digital scholars. And we stopped making the cynical arguments and it started to like anything else when you become excellent, when you become obsessed, when you sort of immerse yourself in it, finally it starts to shift the way you see the world, even your present, the structures, right? I now see in data structures. Like you could talk about the structures of race in America or the structures of labor in America. I see the data structures of America. I see the matrix. No, I see these data structures, right? And how they pattern our world and how it guides people into this box or this box and the bureaucratic consequences and much graver consequences of all of those little decisions that go into how we organize our world, our information systems. Anyway, Claudia went a different way. He went with mapping. And so this was, I would argue, our first successful project. I'm just gonna show you to you on YouTube. And this is how America took over in the eighth of the world. Basically all he did is he mapped every Native American land session from the start of the country to 1887. And it tells you what you already know, right, that the ground that you're on belonged to somebody else once upon a time. And yet, by telling it this way and in a format built for social media, right, a minute's worth of history, this thing went around the world. Everywhere Google does analytics, we had hits on Easter Island. I don't know who's watching this on Easter Island, but I loved it. It plugged into the debate that we were having at the time over who is an American anyway, right? And so it had a contemporary book. It had present value. It wasn't presentist. It's an historical argument that's totally valid, but it did have present value. And so we got, I mean, this is 274,000 views just on our YouTube channel, which is a joke by comparison. When it ran in Slate, it crashed our server. The Slate version is actually better than that. And there are tech people fit in it today. Anyway, so that project was an unalloyed success. And I think it is because it is form follows function. It married the format that it was in to the work that it wanted to do. Open the door to beautiful, light, light of science, blue, imagination. We have a preference for the new high-fragments most in the future. It's awesome. This is somebody's advertising on CSID. So Claudia went that way with the nothing. And I came back to this little piece that had worked, I thought, about Indian nation, the part that we're about to quarantine or destroy, which was the Drupal CMS. And I just, I got into Drupal enough that I could DIY it. And that was so empowering. Not to wait for tech people, not to wait for money. That I can't express how fun it has been for me. And you know, sometimes frustrating, but it really hasn't been that hard. So CSIDXE takes the coroner's inquest from six South Carolina counties. I started with six, all of the photographs on the site were taken by my wife. And that's my backyard, because the origins of CSIDXE are at my house. But here are the six counties that we started with. County level records are really sort of the undiscovered country. They're really imperiled, quite frankly. So counties are really the political level that most 19th century people lived on. They didn't live at the state level. State level law really never affected them. But county level stuff, the sheriff, the coroner, the county courthouse, and those were the constituent political units essentially of 19th century American life. And those records remain in county courthouses. So if you can imagine being along here, going into the middle of South Carolina into a county courthouse saying, hey, do you have any death records from the 19th century that show white abuse of slaves? Not good, but they're there. And so the weird thing about the coroner's office in the 19th century South, during nullification comes later. So if somebody killed a slave, probably that person's wife, they're gonna get away with it. And if they'll never go to a jury trial, and there won't be no trace of the record. Not true for the coroner. The coroner has to make a determination because if there's nothing else, this could be a property dispute, right? There's the value of the person, not as a human being, but the value of the person has nothing. And so the coroner's office is going to make some pronouncement. In the case of South Carolina, it's 12 white dudes with the coroner standing over the body of a dead person. This is, at first I thought we would do this as a mapping project because the inquest has to take place where the body lies. It really didn't tell me anything plus their idea of, you know, geographical information behind smiley still. You know, it's like I like to find smiley still, but it's probably not gonna happen. And so this is what I mean about the maturation of my thinking about form. And I'll come back to this at the end, but I hate the word project. I am so dead sick. There's no word that has been asked to do more inside academia and especially inside DH than project. Right? And that could be anything. So it's our taxonomy. It's an embarrassment of our vocabulary that we don't have better terms for what we're doing. So I invented the term for this. It is an archeograph. Hasn't caught on even that part of it, definitely not. But it is an archive, yes, the way lots of projects are, but it's not the work of a librarian. It is the work of a scholar. It has an argument. So it is an argument bearing on its back, the archive out of which it is constantly being written and rewritten. So in this case, I'm go here. So here's the sample. And this is what I mean about the work of a scholar, right, is to take a set of records under review, but then make it in a judgment, in the case of an historian, a historical judgment about those records and to connect it to an ongoing contemporary argument inside your discipline. This is scholarship, that's my sine qua non for what scholarship is. This is the work of a scholar and a work of scholarship. It should count, I don't need it, but it should count toward tenure. And anyway, so rendering judgments, right, is what scholars do in part. And I heard that every judgment that you make along the way in a digital project is part of your argument. So I'm arguing about the violence of this particular place in this particular time and bothering to compare it to nodding and limping. So these are as a percentage. So we're talking about 33%, one third of all of the cases under review are either homicide or suicides. Does that necessarily say that that's a violent place? Maybe, maybe not, right? If the coroner picks and chooses what they review, every woman who died on a birthing couch happened too many times. They don't bother to call a coroner for that kind of thing, right? And so it really says what the society wants to investigate, not what happened. It is a violent place once you get inside all of these. So I have an argument to make about homicide in the Old South. So you can tell my irreverence, all of these were books, they lived in an Old Testament world. So all of the menu items are books in the Old Testament except for Revelation. But here's the short argument that I'm making about homicides. I tried to make it look like a literary magazine essentially. And I break this down by weapon type. There are 323 total homicides in the Old Sample. I'll show you something about an example there. And what they're using, firearms, sharp instruments, guns, etc., not other unknown. And then here are all, here are all 324 homicide requests. This is Aaron, a slave murdered with a chisel. Upon their oaths do say that the said negro Aaron was feloniously killed by a stab on the left side of the throat with a chisel, about one inch and a half wide by the hand of some person on there. That phrase upon their oaths do say became this kind of rhapsody for me because I knew somebody was gonna go out of the world. And as somebody who thinks that historians don't really understand that what they're doing is rubber-necking at other people's mortality and figuring their own and investigating their own and trying to find meaning for their own. I think we don't admit that enough that what really makes the drama of history is that we're all dying. And so this was like mainlining death. So if I'm kind of addicted to, if I'm in a story and a mortality, it's sort of addicted to the stories of how people go out of the world. But yes, I hit bed, rock, paper. If every little tri-folded bundle, I know somebody goes out of the world. I'll know nothing about the noon of those lives, but I'll know everything about the midnight. And this phrase is always in there upon their oaths do say, and then I watch somebody go out of the world. And then I picked up another one. And then I picked up another one. And so what I wanted to do at CSI Dixie was sort of reproduce that feeling and put it in an architecture or in a form or mode that just seems off and a little disturbing. So the only hyperlinks on the whole site are people dying. So inside the text. So right here, for instance, the last words of J. Edward Sims were typical, shoot you damn cowardly son of a bitch, or take this point in exchange, Tom Rutland firing, I will kill you, you son of a bitch. You have a really impagant bleeding you have already. Those are direct quotes and the links are to the actual inquest itself where you can read exactly what happened. But the point that I'm trying to make about form is this is an argument too for me, this aggregation, this listing. There can't be a reckoning without a record. And I think that's what historians provide in part is we count the dead, we give them names. And if justice doesn't come tomorrow, it comes someday because of the work that we did. So there's an inquest where there was an enslaved woman who supposedly died of apoplexy. That's what the inquest found. But in there is a slip of paper from testimony of her daughter they hit mama with a shovel. So no justice will ever be done that woman that we would recognize as such. But naming names, this happened, that person did it. I know you, forever. There is a reckoning there, not justice, but a reckoning. And so that is the motivation behind this entire site. So lessons learned. You need somebody who's totally obsessive. Lesson number two learned that I'm gonna talk about more if you're gonna DIY, I recommend Drupal. And this, we made a mark with this one as well. So we got it right up in the times. We got it right up in Slate. I got additional money from the American Council of Learned Societies to extend CSI Dixie, which is what we're gonna be doing next. Not just adding four more counties, but adding a real data vis component to it. So I feel like the process of determining my forms led me to Drupal, so I'm like, done with CMS. But now I have to think about, okay, what if I need to map? What if I need to do data visualization? Now I'm opening this whole box again about how to marry these things. Because this is unacceptable. I did that in numbers, I think. And you can't have, I can't be adding new cases all the time to my database and not do a live count of my categories. That's simply not gonna work. So we're not absolutely sure what we're gonna do. I'll go through this really quickly. But I've been playing around with click. So here's our corners and quests after the ACLS grant. So we were at 1,583 in South Carolina. We've added 24,000 now from Missouri. These 1,267 from Virginia are all African-American. So there's what we were looking at with South Carolina. And one of the things that I love about data visualization is for me. Like what does it tell me about my own data? So I can see it for the first time, regardless of whether this ever becomes public facing and I think it will. It's helped me ask new questions. And I think this is a critical difference from what Cleometrics, which might not be a word that you know, but it was in the 1970s there were these quants, essentially, among historians who acted like we could answer questions, definitive, old questions, definitively finally and forever with numbers. And it was hubris and it was disastrous. That's not what DH in history is anymore. It was about using data to see a pattern to raise a new question altogether that may yield to old methods. Go to the archive, figure it out, ask smart questions, write well. Find people who embody that historical experience. And so this isn't that surprising to me that the inquest is about violent death. The inquest is about people who've gone out of the world suddenly or strangely or they are strangers. It's not dying at home quietly and that's what men do in that period. Essentially, we could play with this forever, but I won't, because I do want to sort of get to your questions. This is sort of interesting if you, I don't know if any of you have played with click before, but I've taken us, these are all the inquests in the sample to the 1830s. And if we were looking at homicide alone, there we go. All the way up to the 1850s, the cause of death in most homicide is actually blunt instruments and then other, and then firearm, right? And then very suddenly, 1860s should be about the same. We were only at 1830s. Okay, 1840s, we're just crawling up. We're still at blunt instrument in others. 1850s, we're still at blunt instrument in others. 1860s becomes firearm. Something happens, I think, in the South after the Civil War. People load up. Reconstruction is really the story of guns in America and guns that were marketed to the South. Before the Winchester was the gun that won the West, it was the gun in the Klan, the gun that won the South for redemption. So there are all kinds of new questions is what I'm saying. Once I could actually see my data, I was an historian, an old school historian, but I was able to ask those questions cause I could see so much more. All right. Just, I don't wanna belabor the point, but I've been having so much fun with Clifton Down. I'll just give you another example. So I did the US Mortality Census of 1850. Has anybody heard of that? Nobody has. I call it big, bad data. The Census is the big data project of the 19th century. We don't really appreciate this. It is driving the data industry. So the 1850s Mortality Census is a total joke. Here's how they collected it. First public health survey in the United States. Knocked on the door and they said, did anybody die here in the last year? And if so, of what? This is the most asinine data in the world. First of all, Farmer Ted is not gonna say, oh yeah, I murdered a guy last week. Is that relevant? Or my unwed daughter died on the birthing couch yesterday. I don't know, I'm gonna go shoot that bastard of a non-husband. Who's gonna share that kind of information? Farmer Ted also is not a diagnostician, right? So even if they are going to give up information, they said people died of herpes. That is one bad case of herpes for the people. People died of menstruation. Oh my God, that men in that period, medical men in that period are obsessed with menstruation. I don't know that has radically changed. Unbelievable. So they're dying of menstruation. They're dying of masturbation, apparently. Which is a what a way to go sort of scenario. And unlikely. Unlucky. But, so it's bad data. And it all goes up to Washington DC where they put it through this kind of medical wash and they don't really know anything either. And so it's appallingly bad data. I understand that. By 1890, the Montana Census is tabulated by the American Tabulating Machine Company. That's IVF. By 1940, you have, you know, our Center for Disease Control. You have the National Institutes for Health. This is a system that got big, better, fast. And the big bad data, the bigness mattered more than the badness. The bigness drove that data industry. So as bad as this is, it has an important role in how the data got better. The other thing is, it tells you actually a lot about what I call vernacular death. What the people thought they were dying of? That's interesting to me. If that mother thought that her son masturbated himself to death, that tells you something about their attitudes toward masturbation, her thoughts about her son. Same with the question about menstruation and otherwise. So the mortality census was totally ignored because it is bad data, but I think as a historian, the story of how data gets better is an interesting one by itself. So we did all of this work. And we get to see it for the first time. You know, my beloved quants in D.C. who never got to see this, I feel a kinship with them. Like what they could have done or might have done if they'd been able to see their data. And to see the 43,739 unknown, it's not an auspicious beginning. If we're looking for specificity. If we pick a cause of death, let's just go, let's go right smarter. Oh, here's your lone New Mexican masturbator. So they tried to catch all of these in D.C. I want to go investigate this one. So they tried to catch all of these in D.C. Because the people in D.C. were like, I don't think they masturbated themselves to death. One New Mexican masturbator. Okay, yeah, here we go. So these were total deaths by state 224 murders in the year, roughly June 1849 to June 1850. And this is by state. Well, that doesn't exactly tell me anything because per capita, right? A bigger state would have more murders. So I can just go here. And now we're doing per capita by state. The most violent state for murder in the United States in 1849, which is interesting, is California. It's the wild west. But after that, welcome to the south. Texas, Florida, Arkansas, Georgia, Kentucky, Louisiana, Alabama, D.C., on and on. You see the same thing among deaths for worms, which are massively African-American. I don't think you die of worms. I think you die with worms. And so we don't really know what this is except that it's a marker for blackness. So there's this weird thing going on in the data, like what you're allowed to die of. So blacks are not allowed to die of yellow fever. They're just not allowed. They don't get categorized as dying of yellow fever because the whites don't believe they can die of yellow fever, which is kind of important to the whole plantation regime. So they put them in Colorado, put them other places, doesn't look anything like yellow fever, but they're just not allowed. Anyway, so worms by state, Georgia, Mississippi, Louisiana, South Carolina, Alabama, Virginia, Carolina, Tennessee, and on and on. Now the one thing that I find interesting, and this is about how you can ask new questions, if we go to not the South anymore. That's told us by state, but here we are, per capita, that's by state, New Hampshire, Massachusetts, Maine, Vermont, Connecticut, Rhode Island, New York, Iowa, Texas, you have to go a long way before you find a Southern state. So I now have a real question, right? Do Southerners not murder themselves? They seem to go to murdering other people. Did they not murder themselves, or did they not talk about it? I don't know, but I've got a question now that I didn't. It's just for kicks, I can show you that we did. You're able to obviously tell now that I am obsessed with death records, because we went ahead and did Athens. So these are the death certificate, the first public health survey in the United States is 1850, and you don't get a death certificate until 1919, roughly, which is a physician overseeing federally required. There's a little more medical rigor in that system. So if it comes up, it's fine. Again, one of the things that I'm working with is life expectancies in Athens in that period by race, especially, and gender and occupation, with all of which you can get from death certificates. It's not surprising, right? African-Americans live 10 years fewer, and everybody dies early, so 40, right? Is the average age at death? And really, that is the most important thing that ever happened is that between 1840 and 1940, life expectancy across the world doubled. It is the most important story that has never properly been told, because everything else follows on that essential fact. But we'll move away. The only teaching technology that's never betrayed me is chalk, and it's just a funny fact. Okay, and I'll rapidly walk through our other projects because I don't want to run out of time before you guys can ask me some questions, especially about back-end tech, because I'm happy to have that. This was our next project we did as a collaboration with GTRI, Georgia Tech. They had gotten, I'm not kidding, this thing is called Diamond Dot, which might as well be called the Eye of Sauron, as far as I'm concerned. They got like defense department money for this rapid data mining, I mean extreme data mining. And I'm like, that's terrifying, but they need to put it to some good uses too, and uses that they could share with anybody. And so they agreed to help us put together what we felt lacked in the chronicling America, newspaper data set. So chronicling America is the Library of Congress that's millions of pages of newsprint, but you can't, your research results don't come back in space. They come back essentially as a list. And so we wanted to privilege a map. What you first notice is now you know where your holes in your data are. These are states that have giant blanks. There are no records that will come back from that. And if you search on chronicling America and you don't know that those, you don't know really what this sample is that you're pulling out of, you can make some enormous errors or assumptions and not really, oh, nobody said that in Maine? No, there are no newspapers from Maine in chronicling America at present. But there are interesting things. This is usnewsmap.com that this will do. Interesting to me anyway. So Scala-wag is sort of a pejorative term for whites who are voting Republican in the South during Reconstruction. And it's a neologism. It's a word that got invented essentially in the 1860s. It had existed for time out of mind to talk about scrawny cows that you should separate from the herd and shoot. And so when they're calling these whites Scala-wags they are really suggesting that they murdered them, that they're kind of mongrels in their weaker race. There's all these kinds of things that are programmed into it. So now your search results come back and you can see the term flat lines until 1868 when they needed it. The word gets invented when they need it. The same thing's true. I'm looking at just this little, we have a little timeline right here. So that's 1868 is the high point for usage of the terms Scala-wag. All of these, by the way, return the PDFs of the newspaper articles that use the terms Scala-wag with Scala-wag itself highlighted. I'll give you one more example. Another term that needed to be invented to terrify voters. Again, it gets invented in 1864. And all of a sudden all of these places are talking about miscegenation. So we actually won a prize from the Library of Congress for the U.S. News Map. Claudia got to go meet our senator. What a privilege. I shouldn't sit down too. So, but now look what's happening. For four years we were getting nothing done and all of a sudden we have our forms format figured out. We are creating reproducible forms. That first thing that we did, Indian Nation, an expensive one off, was $25,000 and it's going back on the shelf. Now we know what our theme looks like. We know what we've answered the question of what technology we're using for our CMS. The only question now is, oh look, I can build containers. What do I want to build a container for now? So private voices collects the letters of transitionally literate Civil War soldiers. So there are people who would have written the word chair as C-H-E-E-R. That's the way they say it's a cheater, not a chair to them. So they're accidentally phonetically telling us what they sound like and how they pronounce words. And they are also, from an historian's perspective, all privates, all poor. If you want the story of the Civil War's common soldier, truly common soldier, you can't go into the Southern Historical Collection at UNC which was created when J.D. Darula Hamilton dove around in the 1920s forward and went plantation to plantation. Hey, do you have any of your daddy's old letters? You built a monument to elite white families that you then tried to read against the grain, but you know, the archive, that grain is already built in there. So what you have to do is build your own archive. In this case, I call it a haystack of needles. Go into all those haystacks, pull up the needles, give me a haystack of needles that then I can search on to represent their world. And that's exactly what this does. So private voices now has letters from Georgia, North Carolina, South Carolina, Alabama, Pennsylvania, Ohio, Indiana, Illinois, I think. So we can talk about, we're starting to, kind of a mapping function that can suggest how, who uses the word daddy when referring to their father? Or like Dixie isn't used in Dixie. This is how Dixie is what Northerners use to refer to the South, but the South doesn't use it for a while. We have some of that on here already. So for instance, here is, here who is, these are the people who are using baked beans. I guess this is not that surprising. So we're mapping the individuals who use the term baked beans. So when you're mapping the language, you're not mapping where the letter came from, but where the letter writer came from, because that's where the language came from. That was done by hand. There are hundreds of them on the site. Every one of them done by hand. And they were done by linguists who worked really, really hard on this. But when I saw it, I was like, oh no, no, no, no. We can't do that. Every letter writer came from a particular county. Every county has a county centroid. We're gonna map on top of that. And that'll make a lot more sense. But all of the letters are in here. So, let's go to, so here's every letter written by William H. Brighton, for instance, that's in the sample, the fourth. So there's the content of the letter complete with his breaks. Usually they're a footnote, but not in that case. But if you wanted to see, okay, here are all the letters that Nancy received or authored. It's relatively easy, relatively quick. And I can take you behind this one as well to talk about how we did the CMS, but it's just not worked. If you can design a good spreadsheet, you can get it into Drupal. And once it's in Drupal, it's a matter of, this is what's called a view. You expose a view, which is to just say, take my spreadsheet and give it to me in a format that has hyperlinks in it. It is public. It's really not, it's not as complicated as it might seem. And so now we're off to the races. So this is a project. We're gonna build a CMS for arrest records for 1900 New Orleans around the Robert Charles riot. So the six weeks around the Robert Charles riot in 1900. We're mapping arrestees addresses and where they were arrested so that we can talk about patterns in race and policing in 1900 New Orleans. We already have the data set for six weeks surrounding Robert Charles. We've already done all the latlongs. So this should be spun up pretty quickly. We've got another project coming online now. This is the, I think it's 12,000 teachers who worked with Friedman in the South. And so we're going to map them, but also hopefully do what we tried to do within the nation, which is go to schools and get them to care about and reclaim these people who work so hard. This is just a shell, but what I want you to see is how much faster we get, right? Because once you've figured all of these things out and you start to see this way, it fundamentally changes at least, it's changed for me how I do scholarship and how much better my taxonomy's gotten and vocabulary and kind of concept for when I see a box of records, I can ask different questions about how it wants to live in the world. Is this a research project? Is this a pedagogical project? Is this a discovery tool? That is, it's facing toward me so that I can mine essentially what kind of data this is. Is this activism in the case of the invasion of America? No, I need to deploy this. It's a scholarly argument, but it's gonna make an intervention of a different kind if I package it in a particular sort of way. And so that's what I feel like I've gotten better at is from a scholar in a lone wolf model who sees the world in terms of articles and books and articles and books and articles and books to becoming someone who sees the architecture of what something wants to be based on what it is, is data, based on what audience I could find for it, or based on, in the case of CSI Dixie, it is my personal obsession. So CSI Dixie, I've now written part of it as a graphic model. I'm working on a podcast from it. I'm trying to recreate the sense that it's got an event horizon that I have not been able to escape. It is my black hole. And it just needs to be reproduced and reproduced and constantly evolving and changing forms and none of them are quite right because none of them can quite grapple with the misery that existed in that first fricking box when I went there. Just that darkness I can't stop exploring and I can never get it right and that's okay too. And so there is something this might be relatable and we'll end on this so we can have some questions. There's a performative part of that too, right scholar has a, this as performance in a way that all the forms that it takes and they're never quite enough and we just keep going around and around with the fact that it never ends, right? The book ends and it's that forever. And this, it's like an itch you can't stop scratching and then you can't stop picking it. And there's that quality to CSI Dixie too. I don't know where we are on time. I'm sorry that I just downloaded but we wanted to start from some place so that you could see how an historian, not a great one and not someone who's figured out all the answers and went at this for the last eight years. So I'd love to hear what you guys really are interested in and how I can help. Yeah. So for the US News now. Yeah. The text, the searchable text, how does, again, it's from the Library of Congress. Is it just like OCRD and already corrected for you? Just ran through the e-mail. Yes, yeah. So the United States Congress made that available. We just built a tool. So this is what, back to that question about technology and taxonomy and categorizing reasons. I would call US News Matt a tool. It is not a work of scholarship. It is a work for scholars to return results in the one privilege face or want to think about it in a particular way. So how does a neologism spread across? Because you can actually animate this and watch it. So how do ideas propagate in the ecology of news, reprints, how are that, reprint networks. You can map that kind of thing which you could never do from the prominent American side. So it's very tool to do it. It isn't what it's got. But you're absolutely right that without the work that Library of Congress has already done, there's just millions of pages that they've obviously are. And so we just built a little widget on top of it. Yeah. So in the case of the CSI DC, you collected the data. Correct. And in the other cases, most of the data was already collected. Yes. So what's the average time that each project could take and a lot of people working on it and the selection of the tool, digital tool like click sense. Yes, yes. You could talk about this. Yeah, absolutely right. So CSI DC, all the work is in the transcription of the inquests and readying the data. Once the data is ready to go, and we can talk about this in Drupal, you have to write a little script. This is the only scripting that you'll ever have to do. And they have a module called migrate. And really all it's doing is telling the machine where to put. Let's go ahead and do it. So we're CSI DC. So just so we can learn a little Drupal. I'm already inside the admin because you can see I have the admin toolbar there at the top. The way Drupal works is you start by creating a content type. So let's say that I was heavy into a frog. I just love frog sightings in my backyard. You create a content type called frog. And then you would create, we're gonna add a content type called frog. And then I'm gonna give that frog some fields. So this is color for frog. And then what kind of data we're dealing with. I'm gonna call it a list. Doing this on CSI Dixie is gonna be a little harder because I have the permissions locked down because it's my baby. But I could do it on anyone. So you add a new field color. You say what kind of data could be an open field of a large text block or a small text block or just a list. In this case, you would then get prompted how many colors are there. You're just right. I don't know if these are patriotic frogs. Red, white, and blue. Those are the kind of frogs. Then when you go to content, you can add content of type. Frog. And all of the fields that you've created for frogs will be in here. So this would be your data screen for people that are inputting the data or you could go straight from Excel. So you've got a database called frog and then you have a column called color. All of that has been disciplined and cleaned. Then you can use a migrate script to just suck it all in. That part's fast. That's what I'm saying that I like about Drupal is that if your data is clean and that can take forever, as you guys are saying, but if your data is clean and your architecture is smart, then this part's easy and free. Yes, I'm trying to understand a little bit more about the Indian Nation project and kind of what happened. So you sort of broadly went through the issues, but can you talk a little bit more to that? So for each of these data points or each of these people from the census, you input all that census data, right? And then is it actually like Wikipedia? Like folks are meant to go in and individually input their information? Yes. And so in the language of Drupal, a person is a node, but so is a community. And so people belong inside a community. So you can click on a community and you can see all of the people who are in there and write the biography of the community, which is our thought as to what was the constituent unit that allowed them to survive, to maintain their Indian mess was the story of those communities. But it is also a database of individuals. And so yes, the model was that we were gonna collect oral histories, but also just data about these individuals for whom ancestry works much better with certain populations than with others. And so I don't know why, honestly. I think we didn't put the work in. If I was gonna be frank, we had obviously we had Native American partners and we had folks who were excited about this. We weren't in the schools. We weren't on the ground. We weren't doing this every day. It wasn't our obsession. We were guys who could do it because we thought it was worth doing. And I think it is. But I don't think it fell apart necessarily conceptually. Maybe, maybe. But it presumed an audience. And you can't do that. Shouldn't do that. You're right. And you were also speaking to the sort of the expense that went into it. So right, is that a factor? Also is that still a factor? Is it still a factor to maintain? No. Yeah, it is a little more to maintain, but once it's built, it's not that bad. I will say that good hosting, you pay for it these days. The internet is a terrifying place. It's no longer hackers, right? It's robot hackers that are just hammering every website. Drupal is super secure, which was why White House.gov would be on it because you could imagine somebody wanting to do something there. But even like we just survived Drupal Geddin III, which was a hack on all Drupal sites. And so there is security issues that the DIY part isn't as good. Like, as a scholar, and I can, as I say, I could, we can pull up another site or maybe if you contact me, we can talk about how you can do this pretty quickly. But then you need a pro when it comes to the hosting and security and patching and stuff. You need to work with your libraries because otherwise the NEHs and ACLS and those folks won't trust you for good reason because you gin this up in your basement and they know it's gonna get hammered and destroyed. And so they're not gonna put any money into it. So that's where that partnership is really important. But I like that part of DH is my favorite part, the collaborative part, that we treat librarians like actual collaborators as opposed to people who, you know, as opposed to our workforce or point stuff, you know. It's like, no way. They have mad skills that we don't have that we didn't always appreciate as scholars. And so working in teams has been really delightful for me who doesn't like working in the lone wolf model. That said, you can't have too many cooks in the kitchen. We had a project that was fantastic. It's called People Not Property. It's another county level record very much in distress. It's a deed. So back in the 19th century, you had to register things, big things that you bought. Like today you would have to register a boat or a house. You have to, including people. And all of those deeds are also effectively bills of transfer and sale. So the idea was if you could scale this up and get them all, you could map the second middle passage. So the million men, women and children sold down the river to erect a cotton kingdom. You technically could do that. The courthouses are doing today's business and are in massive disrepair. Those records are being eaten by rats right now or dying of some fungus. So it was just such an enormous multi-state undertaking. It needed an act of Congress to make that thing go. And so it fell apart. And the time was, in fact, your question about time, some of the meetings took forever. You had 20-hour meetings on what we'd like to do with it. And it just didn't help. And so sometimes there is a part of that where you need it to be, your team to be, lean, mean, obsessed with just doing this until it's done. Otherwise, it's back to the nobody's baby problem. That's the part of academic work that we don't know. The lone wolf model does have, it's plus side. It's like, you grind it out. So just on you, or this doesn't. I think there's value to it. The scholarship that gets created that way, the community that's created, right? The process, not the product. Especially when you're working with students. I think about that a lot. I think the product sometimes that we create in my classes in DH, product might be garbage. Process was amazing. They created humanity's knowledge, right? So that when they become state legislators, and I'm doing it again, state legislators in Georgia, they don't think that humanity's knowledge was created by that asshole who stayed at the lectern and hammered them for hours. It was created by them. So that alone, even if the product is garbage. And it's iterative, so it'll improve, right? Class after class after class. They're doing the work of today. So, thank you so much for having me. Oh yeah, you're welcome. Yeah. My students next semester, or spring semester, are using EJI's, the Community Remerence Project with the lynching locations. Yes. They did the soil samples. Right. They go access the records, collect physical clients, and then archive them. Yeah. Absolutely have to include some of this. And I'm wondering about, and this is something I can maybe transfer back to the students, is questions about access of data. Any, because a lot of these counties are very fraught with political team, right, still. So I'm wondering about access of data. If you've run on any issues or resistance. Oh yeah. So CSI Dixie, I was, we had a formal partnership with the South Carolina Department of Archives and History that we no longer have. So a very conservative state legislator. I don't. Absolutely not. Yeah, I don't absolutely know who complained that the South Carolina Department of Archives and History could not be associated with CSI Dixie. And you know why. But you know it's true. You don't like what we're digging out that doesn't mean it ain't so. And so, when I talked to them, and they said that they needed to pull out a formal sponsorship, I was, my back was up immediately because like these are public records and I brother. And they were like, oh no, no, no, you can come get whatever you want. It's just that our name can't be on the side. It's just too politically risky for us. So I haven't. South Carolina, I picked South Carolina because county level records in South Carolina cares about its history, not always for the right reasons. So they did a centralizing call for all records over a certain age to come to the new palace in Columbia and the South Carolina Department of Archives and History. But counties, they don't, they've never done acquisitions assessments. They don't know what they really have. And so CSI Dixie began as CSI Kershaw. And I went to the Kershaw County Courthouse and they were like, oh, we don't have any of those. We send them to an SEDH. And I was like, well, what are those? Corners in question, yes. That's exactly right. So those access questions, county courthouses are tough. They're doing today's business is meeting out, punishment of people. And so there's not an archivist there. There's not someone probably who has even an understanding of what they have, why they would value it. And it's too bad. I went to the county archives here, Bratton's Clark County for the first time. It's, you could write eight dissertations out of the records they have in there. It's insane. County level records are the undiscovered country for historians because people live their lives at the county level. And there's just so much there. But you go to other places and they're using boxes and records as a doorstep or they threw them in a dumpster along the way or Sherman burned them all because he loved burning county courthouse. But yeah, access to records and resistance, it'll be there depending on what the inflection of your project is. More so than we're used to. We're used to going into an archive where they're pros and they don't have a skin in the game when it comes to the argument that you're making. Where virtually all the arguments I'm making, the person sitting across from me probably at that county level doesn't work at all. That's a reality. Two questions. When you got towards the end, you had this, you were just explaining how Drupal was finally something to really work with. So one simple question I have is, do you feel though that Drupal actually shapes what, how you're thinking? And do you experience, sometimes, because I have this thing about links, they're always, they're gonna take me somewhere, but what if there was another place to go? I mean, is there some, so in other words, have you thought about how you would change Drupal? Or is there something, I'm not gonna use organic, that doesn't work, but is there something that would be a little more hyper that would shoot you in a direction? Could you, what I love is what you were coming up with was this idea of like it made me question. Right, right. So sometimes I think these systems, like I hate rectangles. Right, right, right. And I'm just a rectangle. Right. You know? So I'm just curious, what are things you wish you would do that, yeah? So I have thought about that a lot, but it's, first I'll say it's also gone the other way. So I think the most important thing that happened to me as someone who was actually DH resistant even when I was doing it, and who's still a little DH resistant now. And, but not for the, oh, it's part of the neoliberal takeover of higher ed. That argument drives me crazy. I will figure that one in two seconds. But this idea that a machine is now patterning what is an organic scholarly imaginative process in some ways. But what happened to me as I read that backward, and I was like, oh wait, they didn't have Drupal, but they had content types. They had essentially fields. This is my century, the 19th century, is the century that datafied everything. And datafication and the structures of information make some things, it just allows blacks from dying of yellow fever. The assumptions that they make that get baked into their information systems become so fascinating to me. And then I apply that today, oftentimes in what the bureaucracy and how a state sees and how a state structures information and how forms are created and formats are created that have these legacy effects. I mean, a fascinating one, right, is in the 19th century, in the 1820, 1830 census, it was the household was the indivisible constituent unit of reality for America. So they surveyed households. That's the way they're thinking of data. America is a collection of households. Once in 1840, they say America is a collection of individuals. To me, something important has happened. Of course, the slaves don't get names and they're on a separate schedule. Of course, the kids are still underneath their parents and they're not equal. We haven't created equality. We've done something really important. I don't think it just mirrors reality. I think it starts to structure reality. It creates a possibility. Everybody's aligned. Once everybody's aligned, everybody counts in some new way. There are new possibilities that are created. So just to go to your point, I am constantly thinking about data structure, but I'm also reading it backwards to think about how they thought about data structure. And then the other answer is, every time I get frustrated with Drupal, I write a graphic map. I explore the same materials in some new way. And so Drupal's version is supposed to be rectangles and links. And what is this unbelievably violent, dark, 19th century information doing inside a 21st century machine? There's something about, I actually want to amplify the justification of, if you read down that list of homicides and it's just like chiseled, whip, hammer, rock, club, there's something about the pace at which this violence is coming at you that I find deeply disturbing. And the fact that it's coming at you in this form and format in a media age that is usually it's leering and death cells, but instead it's coming at you clean because it has hypertext and hyperlinks and you're like, whoa, whoa, whoa, this is a mashup in a way. And so I'm using its rectangularness in a way to kind of create art out of it. And that's why my wife does all the photographs. I'm trying to think about how this same set of records would be seen by a graphic artist, photographer, somebody who's in podcasting, somebody a sociologist. The bizarre part, a social worker would have told me in two seconds what I would have found. Like I did all of that work to get all of that data to see it all clean and he or she would have said, okay, tell me about this place. And I said, well, you got a bunch of underemployed white, poor white men drinking a lot. You have no access to birth control, no access to social services. And they'd be like, okay, you're gonna have a lot of dead babies, number one. There's no abortion, there's no birth control. You're gonna have a ton of water pregnancies. A lot of them are racial. They're gonna have babies down the well, babies in a bush, you forget, that's number one. Number two, your parenting is gonna be awful. So you're gonna have spousal abuse, you're gonna have child abuse, and you're gonna have feral kids that are drowned because nobody teaches anybody how to swim. And that's exactly what you have, is the land of poor whites. I don't think the morgue is necessarily the only place that you can judge a society, but as I say, on CSI Dixie, every society pancerable for its morgue. But a social worker got me right to the heart of what this community looks like before social services, before birth control, before tripping programs, and it was exactly that. Yeah? Okay, stop. How do you use this with students, both in the classroom and as federal student workers who are working on the project? Yeah, that's a fantastic question. I think DH has a positive, there are abuse possibilities in any, so we're creating this system, which I like, which is where we're collaborating, and that's what we call it, with our students. And we're teaching them about project management. Right, for the most part, I always think the United States used to need, the labor market wanted people that could read, write, do a rhythm, take, sit down, and shut up. So that's what our teaching looked like. You will read, write, do a rhythm, take, sit down, and shut up while I talk to you, because that's what you're gonna do for the rest of your life. And now the American labor market really needs people who can create and collaborate in digital space. So I think our pedagogy has to mirror that to some degree. I agree with all of that. And yet, if it's me telling them what they're going to do in terms of labor so that I can then go to the ACLS or the NEH and get rewarded for their work, this looks like we've gotten back to an essentially exploitative system that used to exist at the graduate level to a large degree, right? Way back in the day, you'd have your graduate advisor, you'd work on that guy's project because it was usually a dude, and they would somehow get more credit for it. And so we think about that a lot. Part of it is that they choose everything, right? What project are we gonna work on? Here's, from the records on up, I help them think about how that project might wanna live in the world, and I do have more experience. But at the end of the day, for me, the product isn't a point. That's how I escape that trap. The process is the point. And that's always true when it comes to pedagogy. When it comes to your classroom, you're always thinking about the process. More than did they memorize all of this garbage information that you, the test is really to measure how successful the process is, more than anything else. So that answers just that little bit of your question. I do lots of assignments out of all of these that are much smaller. Just to teach them a little bit about information, information systems, this conversation we've had about data, and data literacy, which I think is incredibly important for this next generation. So we'll use them in those smaller ways where they're not actually producing the project, they're just using it as a scholar. And those are a week long, or two weeks as opposed to a whole semester, which if you're gonna build a project, you need most of the semester. I didn't show them. Those projects live at DigiLab here. I think you've talked to Emily McGinn, or you're gonna be talking to her some more. She heads DigiLab, which is really the teaching arm of DH here on campus. And she has thought a lot, like we have specific contracts when you contract with a graduate student worker to do work on. They have to get credit, they have to get paid, they have to be able to take their data with them. There are all kinds of requirements. The PI has to be doing the same kind of work. And so we have all kinds of requirements that help protect our graduate students. Undergraduates, we haven't scripted that out as a contract yet. I think that might be a good idea in some place where we could go. But yeah, you're absolutely right to be attentive to that. Yeah. Thanks for that saying. Good. And it's really cool stuff. And it brings to my mind one of the things that attracts me to the digital humanities, which is the ability to bring information to an audience that might not otherwise have the ability or even recognize the pathways to get to information. Middle state criticism is a term that's sometimes used. And so I was hoping maybe you could go back and kind of tell the story again and maybe more detail about that inflection point or your group, your team, where it wasn't just that things weren't working for you getting these autographs put together, but it was suddenly something changed where you talked, I believe you talked about present value. But you're also starting to design your materials with an eye to popular television shows. Right. And it also seems to me, and I'm thinking of this in light of some of the readings that we had today, that it's also a question of how you're reaching out to people on social media, who may then be presenting your material, amplifying the message. Can you maybe work, so it's a visual design issue. Yes. It's visual accessibility, it's rhetoric, but it also seems to me there's some really clever to use in terms of some really clever marketing that you either fell upon or figured out. Yes, so now I'll say some of that is a conceptual failing from the beginning to understand that we were doing public humanities. Right, all digital is public, if it's public facing. And so if you're not really thinking, if you're only thinking, if you wanna design a tool or a widget that I call a discovery tool, which is really it's for you, it doesn't have to be public facing. But if your project is public facing in any way, you're doing public humanities, and there are opportunities that come with that that we didn't immediately see. We just sort of turned the machine on and acted like scholars and said, oh, you could do this, but we weren't, we were writing as we always did as an academic and largely for other academics, which is really impoverishing the medium that you're supposedly embraced, especially when you thought we were gonna do citizen history, right? And so, yes, I think that understanding that and building relationships with folks like Rebecca Nunez-Late and didn't hurt and some ties to Silicon Valley, so Invasion of America was on a jumbo TV inside some tech conference in the Silicon Valley. Because our data is actually really cool and really interesting. And if you're absolutely right, if you pitch it the right way, it seems foreign and intriguing and not dusty and dry. And so, we have gotten better at it, but I don't think that we've, what we haven't yet done and we're just now exploring, we shouldn't do that, right? Where's the press? Well, why are we not talking about the press in this, back to the ecology of the scholarly public? What is the press good at? Monetizing stuff, peer review, and exactly what you're talking about, marketing. Why should I, I already had to do the coding. Come on, man, I'm doing more of my share. And so, we're starting to work with the University of Georgia Press, altcive.org, which is the project that I mentioned about the Civil War letters. They're gonna take that on, we're gonna put it through peer review, and then their people think about design, right? University of Georgia Press designs a beautiful book. They can figure out how to design a beautiful website. The principles are largely the same. Marketing the same thing. Marketing of both versus monetizing, much art. But the peer review and the marketing is really belongs there. And so, I would encourage everybody to, you want a small lean team, but you need somebody who's in Informatics, somebody who's Information Science in the library, and could give you the access to the architecture that is secure. Get them to spin you up a Drupal server on a VM slice. That's what you need. And then you have root access to that. If you can get that, you can do virtually everything. There's this admin, people will lock it down for you. And then you need the press. If I can just have a quick look. Yeah. I'm also imagining that you and your team are going to different places than you used to go as kind of old fashioned, non-profit humanity scholars. Yes, okay. To a degree, I've never quite made the transition to call myself a digital historian or a digital humanities person, even though I teach in the Digi curriculum here, I'm just an historian. And I think part of that resistance is that certain knowledge that that label will no longer exist very soon, because we will all embrace, it's sort of like I began my life as a gender historian. Now everybody's a gender historian. Or they better be, or they're a bad historian. It's just part of the way you, everyone sees, not to the part of the way a limited few see. And so I feel like, again, our digital taxonomy and terminology will get better. Where's the digital intervention? Is it in the data collection? Is it in the data analysis? Is it in the digital distribution? It's going to be in one of those places or all of those places I don't know. And each one of those needs a different name. And then we need different names for our forms and formats. So we have an article, we have an article, we have a book and we have an archograph now. Please help me. But as we get better at naming those forms, I really think it'll just be, we won't have this debate about whether you get tenure on this or not. Presses will do peer review for it. It'll get reviewed. It'll be a certain length. It'll have a certain scholarly value, a potency that can be measured in its reviews. And so we'll all be, digital will just be there for everybody to use a little or a lot. And so I kind of stay, I'm just an historian. And but what that limits, and maybe it's just time and imagination is that I don't, I'm not cutting edge. I have no idea where the cutting edge of BH is right now. I just know what I've been piddling around with for nine years. And I have done a little bit of that, not as much as I should have. Frankly. I have a little question since we're talking about peer review and tenure and all that sort of thing. You've talked a little bit about putting it through peer review. Could you talk a little bit more about what would that look like? Right. Well, I mean, again, I wish I didn't have to answer that question. I wish the press would answer that question because if presses are gonna survive, they had better get it together. If we're just book producers and they're just book publishers and libraries or just book warehouses, we are all doomed. Not that I don't love books, I do. But we're not meeting our public where they are and where we can reach them. Social media changes the kind of election that we hear in the United States. You wanna know where your public is, they're right there. And you can go get them if you're smart about how you go about it. But back to the press piece and peer review. I think what we've talked about with the press is more like a peer review after the fact. Lots of people have talked about doing peer review sort of ahead as the project goes before launch. That's too late for us because we've launched all of our projects. I think as we go forward though, we're gonna go with that other model, which is that it's being reviewed at specific stages. Once because we have our forms and formats down, we sort of know that they hit this threshold, all the data's in, but it doesn't matter what it is or the mapping or these outputs. So then somebody would look at that, what I would call the data conception stage, like what questions could this data answer because it is the architecture good, is the data clean. So you'd have an intervention, you'd have peer review there on that piece. And then before you go live, you'd have peer review on this other piece. Improvements that can be made. And then it would go through what, this is my point, is presses really do have a timeline for how to do every piece in the chain of producing scholarship. It's them I think who really need to figure out what the map or the roadmap will look like for a scholar who's bringing a digital project to the market. And I thought about these things because UGA Press will ask me. I'm like, you figured out. That's my usual answer. Because you have to go make those mistakes, really. Just like I made my mistakes on my end. And I think, you know, Mellon and others are starting to put sort of money into thinking with presses about that. I predict this problem will be solved. And there's been a lot of hand-wringing on this question. So I actually think we're reaching an inflection point where the digital, we'll be around the digital turn. And solutions for material problems. That's my belief, too. Yeah. I don't know how much time we have with this. I'm gonna ask a really practical question. Sure. With teaching, so you teach in a kind of history slash digi humanities program. What do you do concretely in sort of giving your students tools for this digit data literacy, which is a kind of a key objective that sounds like in a lot of your projects. Right. What are you doing? Yeah. Okay, so I'm working on a book called Count the Dead for UNC Press, which is really about what happened when we counted the dead. So a lot of people think the most important thing that ever happened in 1840 to 1940, doubling of human life expectancy, people think that that's okay. Well, we got vaccines and we got clean water and cleaning up and medical health. And that's part of the story. Data is part of the story. Our public health system has no eyes until it has data. So I started by telling them, giving them the sense of how important that they live inside of data structures, that their world is, that they're a matrix, that their world is structured by data. And now I use examples like that one I did about households and what it meant in the census when we changed. And so I really use the census and then the mortality census and get us to IBM and the birth of the supercomputer and the punch card and then up to the CDC. And so I tell the story about how important data has been and the structures have been throughout history so that when they go back and they look at a record, they see it like not as an old thing, but as some cutting-edge technology that inquests that little piece of paper, because it's actually a form, we don't see it that way, but it has a discipline to the information that has to be in it. That's why we can data find it. And so once they're starting to see a historical record as something that can be data-fied and that data-fication has consequences of every choice along the way, has real life consequences for all the people who get categorized in particular ways and has kind of blind spots in it, right? Where they got this or that wrong. And so it's really a history of data and a history of data-fication that we started with. And by being able to see how bad these early clumsy attempts to structure a world and discipline it at the level of data, they understand how bad they are and that they're not as literate as they think they are or not as savvy as they think they are about the structures that they live with them. So it's really not that different than what you do to get them woke, right? To get them to a place where they see race or they see gender, sorry. It just strikes me also now that you're bringing this of a lot of the successful projects and the ones that you have, the ones later on, have been dealing with, in some ways, you're dealing with archival questions and bringing them into the digital space. But your Indian nation one was in some ways almost an oral history project that didn't quite take off. It didn't quite take off. And I think that the work I would be doing with my students in some of the theater and performance actually would be more in the sort of oral history realm. So maybe you could speak to what you would do if you could now go back to that Indian nation project and think about how would you bring an oral history project into the 21st century? The first thing we would do is, that's what my model is, that if you build it, they will come. If you go to them and ask them what they want built with resources that you're actually bringing to bear and you're making it a commitment to be there for a long time and in that for the long haul and you're not harvesting their stories to produce your book, right? You've got to go in and earn all of that, which we failed to do. And so I would start with that and then build what they want built. And so if it's essentially a container, I always think of a CMS as a container. Well, what is the container contained? And in terms of fields, what are you collecting? So if you're collecting oral histories, what are the, you're going to need some, if it's two free forms, then it doesn't have enough discipline that you can take kind of core samples and lay it out and compare one oral history to another, right? Even little things like this, just think of this, I believe. That's just oral histories. It's just gives you a kind of a hook so that you can compare or not anybody's familiar with the story core stuff. But I would be thinking about that as like how I can better compare these kinds of things and what I want to see in its totality, what will be the meta that I will not have just 900 or 1,900 oral histories. What will I know or what will I have above that? How does this thing as a chorus, I would be thinking about that too. And how that could be, and I don't know, represented at the level of the data-fication at the level of the structure. But also at the level of the how it's given to the world. It's oral, I mean, to the task of saying, I think you have to go to the website and it just has to be about people talking and about the value of voice. I would be thinking like, I would run up, I would, the podcaster would be my first thought, I would be like, okay, some of these people get woven into will remain the raw materials for other people to create scholarship and create stories, but I'm gonna push a few of them. CSI Dixie does this, so under Chronicles, I write long-form journalism about one case, one sad incident that comes alive in an inquest. And so I'd be thinking about that too, it's like, how do these things combine or how do they get to the public not just their raw state, but another level or another iteration of intervention in their part, creativity in their part. That's just a first blush, I'm sure. Well, thank you all so much, it's been such a pleasure to talk to you. Thank you, thank you, thank you. All right, so what I'd suggest is this works for you, since if we take a 15-minute break and we start again at noon, then we can be done, I mean at 11, then we can be done at one instead of 115 to get to launch a little bit earlier. So I think that we'll, and then we'll pick up after that at the same time. So,