 So, I'm going to start by introducing myself. I'm Nora Dimick. I'm the head of Digital Humanities at the University of Rochester. And for those of you that don't know Rochester, it's in upstate New York. It's a private university, it's a research one. We have probably 4,500 undergrad and an equal amount of grad students. The School of Humanities, which is what I basically serve in the college, is on the small side. So Humanities is not our big focus. We're a big engineering, optics, biomed. We have a big med school. So Rochester is a really interesting place and it's a really, really wonderful place to work. The library is phenomenal. So I'm just going to plug that while I'm at it. And so I do work in the library. I'm a librarian. And I had a chance to, what's the way to politely put this? I had a chance to get into the digital humanities field by kind of never saying no. So University IT at Rochester works on a service level agreement. They do things that fit within their service level agreement. Digital humanities does not. Digital humanities just kind of is out there in the ether and in the cracks. And so when someone would go to University IT and say, can you do this? They'd say no, use Blackboard. And then they would come and see me. So I kind of was a subversive action on my part with very good results. So I'm going to describe a project that I started with an English professor around temporal narrative. I was going to give this talk at the fall meeting, but I was grounded by the ice storm. And yesterday in Rochester I woke up to a foot of snow. So I'm glad I got here. So visualizing temporal narrative. What does time look like? How can we use digital technology to visualize the experience of time? How can we apply it to the study of textual narrative? David Gretem writes, and I quote, the history of textual scholarship is the history of history. At the very moment in each culture that documents begin to preserve their records of that culture, the issues familiar to textual scholars will appear. Inscription, graphic representation, transmission, error, variant, authenticity, and reception. Clearly as the means of production became easier and cheaper, the need to study actual texts became necessary, which was the authoritative addition. The scholarly addition, how did a particular text change? These were the original aims of textual scholarship or what Gretem would describe as a desire to uncover lost authorial intention in the face of what looked like the garbling of authorial composition by unintelligent and meddlesome scribes. With the advent of the computer and the creation of digital text, textual scholarship, which up until that point was hardly indistinguishable from the history of the book, took an interesting turn. As Matthew Kirschenbaum notes in his essay, Tracking the Changes, the computer is a real game changer, and I quote, in the particular realm of literary and textual scholarship, this means that a writer working today will not and cannot be studied in the future in the same way as writers in the past, because the basic material evidence of their authorial activity is like all textual production increasingly migrating to the electronic realm. The digital changes the way we write texts. It changes the historic record of texts. There's no more visible visions or drafts unless authors are conscientiously squirreling them away and not on zipped disks, floppy disks, or in cleris work files. But most importantly, and this is where I bring the presentation back to the topic of what does time look like, the digital gives us new ways to interrogate texts and not necessarily digital texts, but any kind of texts, which is where digital humanities comes into the picture. It gives us new tools to look at texts in a new way. It opens up new worlds of possibilities for reading texts. And you realize this is about television. So close reading. First Professor Joel Burgess would say that close reading provides a perceptual education, one that demands historical, and I would argue, technological context. Close reading as a form of textual scholarship then allows us to see how representation works and as Burgess would say, works on us as a sensory, affective, and cognitive phenomenon. We need a minute for that, or we're good. Now we're back to television. So Poetics of Television. It's a class taught by Professor Burgess, an English class but also cross-listed with film and media studies and digital media studies. The course looks at television narrative within the framework of the mode of technological production and presentation. For example, how does the temporal structure of network television affect the temporal structure of the narrative? This can be seen very clearly when we talk about different genres of programming. The sitcom narrative generally takes place within a 30-minute program, but is in fact only 22 to 23 minutes long. And within that self-contained or closed narrative, there might be flashbacks, flash forwards, but there's rarely a connection to other episodes or a sense of real-time passing. Do Jerry and Elaine ever grow old? The soap barbers an open narrative. The continues on, and the subplots may end, but there's generally a continuity of characters. The miniseries is closed, but not in the same way that the sitcom is. The narrative is much more expansive due to the longer presentation time. The procedural is both open and closed. For instance, in an episode of Law and Order, a character might be undergoing some kind of terrible problem over the whole season. But during that, in each episode, a crime is committed. It goes all the way through the justice system, and it's resolved in a very, very compressed timeline. So Professor Burgess wanted to think about, how can we visualize this? So visualizing TV time. So in order to interrogate this idea of temporal narrative and technology, Professor Burgess and I conceived of a data visualization project whereby students would collect data about television shows. During the time of shot changes, scene changes, and other narrative elements like flashbacks, flash forwards, and diegetic time markers. And diegetic time is narrative time, so in Law and Order, if June 1st, 1998 flashes on the screen, that's a diegetic time marker. So we hacked a couple of tools to make the process easier for the students. So there can be several hundred shots, and you can see there, that's a 22 minute episode of Cosby. There can be like 250 shots in a 23 minute sitcom. So we used a free color grading software to automate this procedure. And I say automate, but they still have to look at it and do some cleanup of it. And then after that, they take a text file, and they put it into a subtitling software that's cross-platform called Jubler. And we also co-opted Jubler for our own use for this. And then they can annotate the shots. Okay, here's a very simple visualization we produce from student data. So if you look, you can see that between 1952 and 2010, the number of shots per episode of comedies increased dramatically. Not too surprising when you consider digital technology coming along. So it's interesting, and it's something that you know, but now you can know it in a different way, if that makes sense, and students really gravitated to this. So here's another interpretation of the sitcom data. And this could be used to tie to an argument about stylistic sensibilities. So if you'll see the number of scenes and beats, the green and the red changes over time for the sitcoms, like quite dramatically. At any rate, the project accomplished its objectives with really good results. Students in the class wrote outstanding papers that tied the theory they were reading to actual data they derived from their visualizations. And in the process, they learned how to analyze visual media and interrogate their assumptions about what a shot is, what a beat is, and what a scene is. In fact, Burgess said that those were the best papers he ever got. Here's another visualization. If you look over there, there's the character network using network analysis of the Simpsons. And over here we have Seinfeld, and it's really dark. It's hard to see those. But if you can see, obviously the Simpsons doesn't have to pay any actors. They can have a million characters. So that was another thing that they could think about and tie to it. So no good deed goes unpunished. As a result, because it was so successful, I ended up hiring one of the graduate students as my DH programmer. And Professor Burgess decided that we could extend our skill set and teach a semester-long class with the curriculum that we came up with ourselves that we continue to probe these ideas of narrative temporality. So the class was co-taught by Professor Burgess, the programmer Josh Romp, and myself last semester in a computer lab so that the students couldn't escape the influence of the digital even though half of the semester focused on a close reading of this book down in the chapel, The Religious Life in an American Prison, kind of daunting. So we started by reading the book over the summer. Joshua Dubler is a religious who wrote the book, is a religious professor at U of R. And we thought it was really great as a project because after we could do the close readings and do the presentations, the author would actually be able to come in and interact with the students and talk about the themes in the book. The book is both an ethnographic study and a non-fiction narrative focusing on one week in the chapel in Gratterford Prison, which is a maximum security prison in Pennsylvania. Each chapter in the book is a day of the week so that the narrative lends itself perfectly to a close reading that is also closely tied to temporality. Like poetics of television, the class is also a mix of graduates and undergraduate students and it was a really big class, 26 students were in the class. So we started this project by bringing in poster board and newsprint and markers and asking the students to start to think with pen and paper to imagine the answer to their research questions after a close reading of the book and what their visualizations might look like without being constrained with the idea of thinking, I have to collect data on this and I have to use Excel or I have to use R or I have to use, you know, JavaScript or something. So this one here, this student was trying to represent the series. So each line is one of the characters and as you can see, as they're talking in the book and as Dubler is capturing their narrative, they talk about the past, they talk about the present, they talk about the future. So he's trying to capture that flow from the past to the present to the future and he has a scale of years at the top. This next one, if you see at the top, they wanted to think about the framework of the prison and how every day was pretty much the same and how Dubler could really write this book Monday to Sunday because every Monday in the prison was the same as every other Monday and he never needed to do more than one week. So they really wanted to capture the prison time, the cyclical nature of prison time. So they ended up focusing on using Excel, which is really easy to use and as you see when it contracts, that's when the prisoners are talking about their past. But as you can see from the visualization, they spend most of the narrative in the present day. So that's a really easy way to see that. So I'm going to read from one of their papers a little bit. The first quote that I have is that they talked about unending days continually folding back in on themselves, unavoidably connected and rotating around the same axis. So that was what drove their visualization. So they collected data about the tense of the narrative. When Dubler or one of the other speakers is talking about the present, it's 2006. We call this story present. Other dates are represented by their actual year. So in the cyclical graph, you can see that the majority of the narrative takes place in the present. Then they took this a step further to explore issues of religious temporality. What does Christian and Muslim time look like? Do the Christian inmates spend the same amount of time in the present? They explored why inmates might want to convert to Islam or Christianity. Thus when an inmate chooses to hope in the form of religion, at least Christianity, he is improving his sense of now and in doing so, perhaps making his time his cyclical, repetitive wasted time more meaningful. This is an apparent in the fact that the Christians referenced the present more than Muslims. So they took this as a jumping off point and really did a deep dive into the religious life of the prison, which they didn't have to do. It was a completely open-ended exercise. So they also rebelled against the subjective nature of the selection process of the tools and concluded, rarely is the goal of humanity scholarship to find the right answer. They were a pretty good group. Here's another one. This group took each day of the week and what they did was anytime that someone was talking about the past, they did a backslash. The future, a forward slash, they counted the stories. So they collected actual data on this. And this is what their visualization looked like. And then I'm going to spend a few minutes on this one because it's a succession of works that built on each other and finally ended up with a really, really interesting outcome. And of course these were graduate students. So this group was really concerned with staying true to the materiality of the book. They wanted to represent it as words written on paper. So they used tracing paper over the book to look at the textual density as a function of how long the reader spends in story past, story present, or story future. And here's another one of their drawings. And if you can see, you probably can't read in the middle, but the middle circle says before the raid and after the raid. So that was a focal point of their visualizations and the narrative for them. Everything happened either before this pivotal action in the prison of the prison raid or after it. Now they've evolved to this starting to think in a timeline and how could they continue to represent that in a timeline and use all of their own talents. Their final written paper was called Loose Correlation, Reading Textual and Temporal Densities. And they write, they were looking for, and I quote, a sense of the rhythm of prison life that was spatial rather than linear, irregular rather than consistent, and anachronic rather than chronological. Here's another one of their early ones where, again, they're playing with page density as calculated by the pixel ratio of the page. And those numbers, I thought, what are those numbers? They're the page numbers. Nothing really crazy here, right? So this represents their textual density, and again, they're still trying to be true to the materiality of the book. They created a loose formula for density of text and assigned a different color gray to it. The denser the text, the more saturated the gray. The numbers are the page numbers. And so they looked at this, they worked really hard at it, and then they said, what does this tell us? Nothing. Because it would look really different if they were reading it on a Kindle or on a trade paperback. So it was too tied to the text. So this is their next attempt. This is how their final visualization took shape. This shows the beginnings of their idea, and I don't know if you can see the black line across the middle, that's the prison raid, and at the bottom, that's kind of like 2006, that's story time. Story present is the bottom, and so every time that they spent time in the past or the future, it either goes above or below those lines. The other thing that they really tried to do, and tried to work into it, was to add religious time, because they wanted to add that and be true to the text. More of these, okay. So they overlaid religious time on top of the elements, so that references to religious events are captured on top, and primarily Christianity, Islam, and Judaism. And so those are, you can see things reaching out into the future. And luckily, this team was very tech savvy. They used processing, a computer language, to create the visualizations of down in the chapel. And if you see the tap row, that array represents the variables they chose for the past. And then they used an algorithm to smooth it out. Originally, they didn't have the algorithm, and the religious dates way far in the past and way in the future were basically ruining their visualization. So they were really concerned with the aesthetic quality of it as well. But they decided that they couldn't just throw the religion out of the story. So it had to be in there. And so they created this algorithm to smooth everything out. This is what they write. The summarizing process of measurement, then, is a process of representation that introduces vagueness into a system of much more nuanced exactitude. Somewhat paradoxically, however, it is this character of inactitude that enables knowledge to arise. As in the case of scientific measurement, vagueness as themselves and among disparate, I'm sorry, vagueness as representation is what makes distinct concepts mutually intelligible, both among themselves and among disparate recipients. This observation would seem to apply readily to Dubler's book, particularly in light of his emphasis throughout on the fog of grays of religious life in Gratterford, with gray being an archetypal example of vagueness in a world more desirably constructed in black and white. In this light, his structuring of the book in today's erects a facade of order and precision only so that it can crumble. So this was an amazing semester. They really, all of the papers were incredible and outstanding, and everybody got to bring their skills to bear in this class. Some did R, some did Excel, some used Photoshop or Adobe Illustrator, someone did a YouTube that showed time passing. It was an outstanding, outstanding effort. And so, again, we revisited Poetics of Television in this past semester with a new student body. And because we were able to support new types of projects and data collection because of our experience in the past two classes, the students could ask different questions. So this is one of the students who wrote, what is the class structure of the spaces in upstairs, downstairs? So they looked at and then they tracked where everyone could go. And interestingly enough, the servants are the only ones who had the run of the house. And so they thought that was a very interesting concept and counterintuitive to what they would have thought about upstairs, downstairs. And of course, they had to do Downton Abbey. And they found the same thing, pretty deeper character network, but the same thing. And they did an awfully lot of work around this, collecting all of the data around this. So 21 is the character code for characters in our data model. And they assigned a third digit to each character. And they really worked really hard to collect this data. And so I'll talk a little bit about that. We do quite a bit of data modeling with them and thinking about how do you translate concepts into numbers because you need numbers to create visualizations. So that can be some of the hardest part for us. Luckily, we have a lot of science students. So usually we have one in every group that's an outstanding coach and is willing to share their expertise with the others in the group. And that's another thing about humanities is having these groups and humanities, whereas the single scholar model is really going away in this kind of work, which I think is really interesting. So we talk a lot about aesthetics and epistemology. And we asked our students to consider questions of aesthetics and epistemology. And how does it help us know something better if we can see it in a different way? And then again, it's the scholarship. What's the scholarship? Is the data the scholarship? Is the visualization the scholarship? Or is it only through their writing about it that creates the scholarship? And so there's some ideas around that that we're really thinking hard about. And how are we capturing this? How are we preserving it? Or should we? Does it have the ephemera of a term paper? And finally, we're thinking about new frameworks, right? So humanities is a great jumping off place for any interdisciplinary collaboration because it asks fundamental questions about what it means to be human. A question that could be explored through virtually any disciplinary lens. They are not without care, however. Collaborative projects require management. There needs to be a clear understanding of who brings what to the project and when. And good collaborations are built on trust. You need to be responsible and dependable and meet your deadlines. Using a project management framework can make it easier to approach deliverables and take the sting out of someone being the task master. Especially if you're working with someone for the first time, clear expectations and agreed upon outcomes and objectives can be the first step. When I first met with Professor Burgess at the start of the project, I asked him point blank, what were his objectives for a humanities lab? I never want to promise something I can't deliver. And so the flip side is to help your collaborators understand the scope and the limitations. We are not going to create the Sistine Chapel in 3D in a semester with a group of undergraduates. Not going to happen. So these new frameworks for librarianship are built on disciplinary collaboration and they allow us to reimagine our practice as librarians. What if we were sitting in on the research team as a data consultant or a metadata consultant, a grant writer, programmer, curriculum designer, copyright expert, digital image specialist, research consultant, fill in the blanks, right? They help us build our capacity in scholarly communication and interdisciplinary research. And we can measure it through course evaluations, enrollments, increased collaborations, publications, and grant funding. Every class Professor Burgess teaches gets bigger. He has a sorority following. They all take his classes. But anyway, he's a legend and he's junior faculty, not tenured yet, so this was a big risk for him and I think the payoff was really great. And finally, our work with the Poetics of Television and clocks and computers was transformative. It changed the way we thought about our work. We were no longer supporting faculty, but we were collaborating and working alongside them to create a project that strengthened the curriculum in a way no study guide or Prezi or bibliographic instruction could. We were designing research opportunities that make the concepts of film and media studies, production, transparent, and literary studies and that could be used as a basis for a different kind of analysis that students and even Professor Burgess had ever done. And then that's what I've got and I'm hoping that we can open it up for a discussion and I can answer any of your questions. Well, if there's no more questions, thanks very much.