Abstracts
Long and Short Presentations
Taking Flight: Expanding a Crowdsourcing Program in Support of New Research
Mark Anderson
Presentation Slides
DIY History is the transcription crowdsourcing program at the University of Iowa Libraries. Starting with Civil War Diaries, a small part of sesquicentennial events in 2011, but that went viral after being picked up in a Reddit post, DIY History has now expanded to 10 collections and nearly 65,000 transcribed pages. It acts both as a public engagement tool, and as a source for class activities on campus.
Most recently, its focus has grown beyond just the transcription of diaries, letters, and other manuscript materials with a collaborative project between the Libraries and the University of Iowa Museum of Natural History to digitize and transcribe a collection of field observation notecards gathered by 19th century amateur ornithologists. The Egg Cards describe specimens at the museum, but having never been transcribed, searching the collection remains a manual task. The results of the project will not only provide searchable data for the collection that can lead to new research and scholarship opportunities, but also draw upon the participation of new citizen archivists. My presentation will describe the changes to the DIY History program as a result of this new project, and ways that it continues to support digital humanities at the University of Iowa.
A Guerrilla Theory for the Digital Humanities
Matt Applegate
Presentation Slides
This paper charts a political economy of the digital humanities and interrogates the invocation of the guerrilla as a means of conceptualizing DH praxis. Featured most prominently in the discourse of the #transformDH movement, the guerrilla, following Natalie Cecire’s description, is a valuation of “the oppositional, the maroon, and the fugitive that characterizes #transformDH [that is] clearly indebted to the legacies of queer theory and critical race studies.” Simply stated, connected to contemporary concerns for postcolonial DH and #activistDH, the guerrilla operates as a means of thinking DH praxis from a position of radical difference. Beyond DH praxis, the guerrilla is invoked in the discourse of New Media and political economy as an ungovernable monster, a pack of wolves, and an active nihilist. It is, parallel to the DH invocation of the figure, a figure made ‘other,’ but a divisive other, one that is both violent and unpredictable.
I follow this duality of signification by exploring collaborative and collective acts of political becoming endemic to guerrilla discourse and tactics. Augmenting the discourse of post-colonial DH, I turn to the heritage of Fernando Solanas and Octavio Getino’s Towards a Third Cinema in particular, arguing that collective acts of cinematic practice in Latin America might serve as an analog for a guerrilla DH. I conclude this paper by theorizing the guerrilla as an inherently collective figure and an expression of what Cathy N. Davidson thinks as “collaboration by difference.”
Digital Text Encoding of 18th-Century French Novels Reveals Modern Misconceptions About Narrative Structure
Benjamin H. Baker
While today we consider the chapters of a novel to be the basic building blocks of its narrative structure, this idea is a result of the way novels are published, not an inherent characteristic of the form. As I will demonstrate in this paper, using digital text encoding to analyze the structure of pre-Revolutionary French novels shows that this misconception masks a complex interplay between narrative structure and dispositive structure (chapters, books, volumes, parts, etc.). In pre-Revolutionary France, novels were often published in installments separated by a year or more, so readers did not expect them to be published all at once, and if the novel was successful even a seemingly definitive conclusion might be followed by further installments, which might not be the work of the original author. When Éléazar de Mauvillon published a continuation of Antoine François Prévost’s Mémoires d’un honnête homme, he modified Prévost’s work both by making additions to the original text to prepare the way for his continuation and by assembling the narrative structure of his portion of the text in a way that reframes the events in Prévost’s portion of the narrative in a new context. Mauvillon’s continuation of Prévost’s text exhibits striking similarities to the modern phenomenon of the “reboot” and its attendant use of a technique now popularly known as “retconning,” which describes the addition of details after the fact to explain apparent inconsistencies. Digital text encoding techniques make it possible to perform comparative analysis of the narrative and dispositive structures of novels like this that have multiple authors and states of publication more effectively, and could be applied to works of narrative fiction in other media that display similar characteristics.
What Made the Front Page in the 19th Century?: Computationally Classifying Genre in “Viral Texts”
Jonathan D. Fitzgerald
Presentation Slides
As part of the Viral Texts Project my colleagues and I study 19th-century newspaper content specifically through the lens of virality. In US newspapers from the 1830s to 1890s, for example, we identified more than 1.7 million reprinted texts, but the clustering algorithm only tells us that these texts were reprinted. The researcher interested in specific genres such as poems, recipes, or advice columns, still must search and browse an enormous corpus to find what they seek. To that end, we have begun to computationally classify these texts by genre.
In order to begin mapping genres in a corpus of 19th century newspapers, I am utilizing an innovative classification method rooted in Natural Language Processing developed in collaboration with Benjamin Schmidt, Assistant Professor of History at Northeastern University. This paper argues that using a method that combines text analysis tools such as topic modeling and logistic regression, a successful genre classifier can be trained. Using topic models as opposed to tokens as the classifiers is beneficial in that it logically limits the variables while also being human-readable.
My paper will show how I utilized this approach, first by hand-tagging a selection of clusters to use as a training model and then by implementing that model to identify genres within the corpus. In the end, this work will advance the work of the Viral Texts Project by allowing us to identify just what kind of content “went viral,” and in what amounts. Additionally, the tools and methods used in this project will assist other researchers in their efforts to disambiguate genre in a variety of other domains.
Augmented Reality as a Platform for Collaboration
Robert Fletcher
This presentation will discuss the potential and demonstrated promise of mixed or “augmented” reality technologies (AR) for several kinds of narrative collaboration: between programmers and artists; between different storytelling media (image, text, audio); and between author(s) and readers, since, in the reader’s experience of AR, the author’s programmed reality mixes with the reader’s own experiential reality.
In order to develop this argument, I will examine a number of different examples of AR, both professional and DIY. First, I will offer a reading of the collaborative aesthetic of various AR practitioners, such Amaranth Borsuk, who has teamed with Brad Borsuk for the noted AR poetic narrative Between Page and Screen (2012) and with Kate Durbin and Ian Hatcher for the intermedial artist’s book (and accompanying performance pieces) Abra (2014-15), and Aaron Reed, whose collaboration with Jacob Garbe, Ice-Bound (2014-16), crosses borders such as narrative/game, book/program, and human/AI. Next, I will discuss my own collaboration on AR versions of 1890s picture poems by “Michael Field” (Katherine Bradley and Edith Cooper), two Victorian women poets who explicitly promoted a collaborative aesthetic that anticipates Borsuk’s feminist vision of digital collaboration. Finally, I will tell the story of my collaboration in the classroom with students creating their own versions of AR narratives and games.
Adaptive Forms: Conditional Logic and Advanced Survey Engines
Mark Gallimore
Humanists and social scientists use online data-collecting forms for a variety of scholarly tasks, and a major advantage of digital forms is conditional logic. Most major survey or form engines have a big set of logic options that allow a form to change as the respondent enters data. Forms might add or eliminate subsequent questions based on a respondent’s earlier input. A form might install information a respondent entered a moment before into a subsequent question, so the subject of that question is clear. For those filling out a survey, or providing respondent-specific information, the form is only as long as it needs to be. This keeps each respondent focused on what matters to them, and gives them a sense that providing data to the project is worth their time and effort.
In this twenty minute session, I demonstrate conditional logic and piping using an advanced survey engine typical of those available to scholars at most colleges and universities. I also demonstrate the simpler but still useful version of conditional logic in Google Forms, a tool available to everyone for free. Participants with devices are encouraged to build a simple form along with me.
Student Videos: Teaching the Mechanics
Mark Gallimore
Please see the video abstract at: https://youtu.be/YK8Zg44EVW8
R for Humanists: A Git Repository and Tutorial Approach
Tassie Gniady and Grace Thomas
The Cyberinfrastructure for Digital Humanities Group at Indiana University—Bloomington has been working on building simple R scripts and RNotebooks describing how each line of code functions that are freely available on github. Tutorials are also offered on campus four times a semester, with an additional two planned for Indian University—Purdue University Indianapolis. As R has exploded in popularity as a way to do to text analysis, we have wanted to make available a repository that is aimed specifically at digital humanists and that may reach those unable to attend summer institutes, or who want a tutorial-like experience while broadening the applicable fields that the pathbreaking book by Matthew Jockers, Text Analysis With R for Students of Literature, addresses. Our first workshop had attendees from ethnomusicology, folklore, history, library science, and Italian & Portugese, in addition to a surprise constituency: epidemiology! We also addressed the benefits and drawbacks of using a library like tm (text mining) for simple analysis versus writing functions on your own. Our more advanced workshop looks at sentiment analysis, which is beneficial to a broad range of disciplines and can be applied to social media, survey, and text. This presentation will share what we have learned from the inaugural semester of these workshops, the ten inaugural scripts, plans for building interactive Jupyter Notebooks this summer, and different methods of teaching programming to humanists.
“What Does It Matter Where One Lives?”: Spatial Mapping and Edith Wharton’s The Age of Innocence
Meredith Goldsmith
This presentation traces my development of a spatial mapping project based on Edith Wharton’s The Age of Innocence for undergraduate literature classroom use. While I had initially believed the process would be one of learning tools and applications, I learned that a digital project would require reconsidering my scholarly processes and my knowledge of the text itself. Using digital tools required me to reconceive a literary text as data, isolating rather than complicating my evidence and looking for specificity rather than ambiguity. As my comfort with GIS tools increased, I began to think critically about them—just as one would with any literary methodology—and not just as a means to an end. A tool designed for quantitative analysis afforded new qualitative questions; although readers can discern the metaphorical confinement of Wharton’s characters from her novel, visualizing the text via GIS made the entrapment of her characters newly palpable. Exploring the spaces literary characters traverse and inhabit led me to view mapping as a distinct form of interpretation that makes humanities scholarship relevant to undergraduate students’ lives. I view this project as a first step in a broader collaborative endeavor using GIS to map literary space across a wide range of texts. My larger goals for this project are twofold: first, to increase students’ engagement with literary and cultural spaces, thus bringing texts alive in new ways; and second, to build the credence of GIS among humanists, especially those might have qualms about quasi-scientific approaches. Lying at the intersection between qualitative and quantitative, art and science, landscape and grid, maps offer an ideal tool for such an endeavor.
Effects of Task Complexity on ESL Students’ Argumentative Writing: Using DocuScope as a Tool for Analyzing Students’ Writing
Maria Pia Gomez-Laich
This study tested whether task complexity, defined as task features that can be manipulated either to increase or decrease the cognitive demands placed on learners while performing a task, affects L2 learners’ writing performance. Specifically, the study examined whether increased task complexity affects L2 learners’ ability to (a) use genre conventions for writing academic argumentative essays in English, and (b) use linguistic resources that are generally employed in argumentative essays. The participants were 26 international students enrolled in an undergraduate writing class at an American university. During the instructional sessions, participants completed one task in dyads. Half of the students completed a simple task and the other half completed a complex task. Although both tasks required students to write an argumentative essay collaboratively in pairs, students in the simple group were given the topic, arguments that they could copy and use in their essay, the rhetorical moves of argumentation, and linguistic forms to be used to express their ideas. In contrast, students in the complex group were only provided with the essay topic. To measure learners’ development of their argumentative writing skills, three individual essay tests were used as pre-, post-, and delayed post-tests. To analyze students’ essays, this study implemented a corpus linguistic approach. More specifically, it used the text analysis software DocuScope (Ishizaki & Kaufer, 2011) to look for trends that serve as indicators of academic writing development. Those indicators include increased use of certain linguistic and rhetorical features prominent in academic writing (Institutional Register, Academic Register, Reasoning, Elaboration, and Reporting) or decreased use of certain features that are more prominent in informal oral or non-academic language (e.g., Description and Narrative).
Toward a Theory of Time for the Digital Humanities
Tim Gorichanaz
The dominant view of time sees time as homogeneous and one-directional. But this does not adequately characterize the human experience of time: Some hours feel longer than others, for instance, and we may forget what day it is. Moreover, we recall and reshape our pasts, jumping backwards, and we plan for our futures, jumping forwards. When documents become involved, the experience of time is all the more multidimensional. Up to now, the digital humanities have been operating with the dominant (positivist, physical) view of time, but in order to provide a holistic representation of the human condition, other models of time should be explored. This paper argues that Heidegger’s theory of time, from phenomenology, and the theory of document transaction, from document studies, can be used to present a theory of documental time. For Heidegger, time does not exist, per se, but rather unfolds as part of being. Being and time are characterized by the fusion of past, present and future; the three are not simultaneous, but they co-exist and can co-determine each other. The theory of document transaction postulates the document as the momentary coming-together of a person and an object. A document transaction is the mechanism by which a document comes to be; thus the document is neither the object nor the person, but something that arises when the two meet. In documental time, then, the past and future of the person and the past and future of the object cohere in a shared present. This view of time invites a host of analytical and visualization strategies for the digital humanities to explore.
Augmenting the Historic House Museum: The Impact of Community Partnership and Augmented Reality on Visitor Experience at Riversdale House
Quint Gregory, Nicole Riesenberger, and Caroline Paganussi
Presentation Slides
In Fall 2015, the Michelle Smith Collaboratory of the University of Maryland’s Department of Art History and Archaeology partnered with the Riversdale House Museum to devise a technologically innovative solution to some of the two-hundred year old house museum’s most pressing challenges: funding, visitor engagement and accessibility. While this partnership promises to benefit both institutions through upcoming courses and other projects that capitalize on the strengths that each institution can offer, the initial target of the collaboration is the bi-centennial celebration of the nation’s first blockbuster art exhibition, which involved the public display of dozens of old master paintings at Riversdale House in 1816 by Rosalie Stier and her husband George Calvert. In preparation for the opening of the anniversary exhibition in April 2016, a team of graduate students and museum professionals has begun to create a range of interactive and multimedia content that we hope will enhance visitor experience and accessibility in the museum. Using iPads and the Augmented Reality app Aurasma, we will present viewers with audio, video, an interactive provenance map, and a 3D reconstruction of the carriage house where the paintings of the Stier collection were stored in crates for several years. Perhaps most importantly, by integrating touch-screen technology into the museum space, we are able to provide visitors with zoomable, high-resolution images of paintings that could only be included in this exhibition by way of printed reproductions. Our presentation will focus on the design, planning, and testing phases of this collaborative project, and will include initial evaluation results detailing the impact of this augmented reality intervention on visitor experience. We seek to demonstrate the ways in which augmented reality can revolutionize the historic house museum and art exhibition, and to suggest future applications of this technology.
German POWs in the Soviet Gulag: Mapping Their Experiences, 1941-1956
Susan Grunewald
There is no historical event so central to Russian identity today as the “Great Patriotic War,” or World War II. My project examines German prisoners of war in the Soviet Union from 1941-1956. The Soviet government held roughly 1.5 million German POWs in the Gulag system after the end of the war, the largest and longest held group of prisoners kept by any of the victor nations. My project centers on the following questions: Why did the Soviets keep the German POWs for so long? Was their incarceration politically or economically motivated? Finally, how did they fit into the Gulag forced labor system?
Based on preliminary research, I argue that the Soviet government detained the POWs primarily as a needed labor source to aid in post-war reconstruction. The war did tremendous damage in the Soviet Union: the country lost 26 million soldiers and civilians and the industrial and agricultural infrastructure in the occupied territories was destroyed.
A significant part of my project is mapping. Using Russian archival sources and the program ArcGIS, I aim to answer key questions related to the German POW Gulag experience including: First, were the main POW camps located in rural or in urban areas, and how isolated were they from the Soviet prison population and the free population? Second, what contribution did the POWs make to reconstruction? Were they primarily used for resource extraction? To what extent did they have a hand in reconstructing the massive destruction of cities? Third, how did they fit into the Soviet Gulag forced labor system? Were they primarily distributed to POW only camps or were they integrated into existing ones for political prisoners and convicts? Finally, how did their geographical location shape the experiences of POWs?
Mobile Media Storytelling in the Cultural Heritage Sector
Chelsea Gunn and Aisling Quigley
Site-specific narratives, from physical signage to guided walking tours, are a longstanding tradition in cultural heritage organizations. More recently, the adoption of locative mobile smartphone applications has provided new mechanisms for art museums to supplement or augment museum exhibitions. In the context of historic sites, these initiatives often take the form of mobile walking tours and audio walks or augmented reality applications that allow users to view historical images of a specific site from within its present-day location. Often, they are described as innovative tools for outreach, contextualizing collections in the physical environment and promoting access to an expanded audience.
Despite its emergence across institutions, however, the uses of mobile technologies within and outside of physical exhibition spaces have not been thoroughly investigated and it is unclear whether and how mobile engagement is occurring. Through two studies, this presentation will explore this problem, investigating the distribution and use of mobile apps and walking tours at accredited art institutions and historic sites in the United States, and their intended and actual use in various contexts. Studying site-specific mobile applications provides a unique opportunity to consider when and where digital engagement occurs. By viewing mobile media storytelling applications through a participatory lens, cultural heritage organizations may reevaluate how content can be effectively disseminated, but also how they are situated within their own communities.
Toward a Taxonomy of Collaboration
Jacob Heil
After a couple of years of team-assembling and project-building under the auspices of the Ohio Five’s Digital Scholarship initiative, we have seen a variety of kinds of collaboration, each of which requiring different technological infrastructures and campus partnerships. I propose a “long presentation” (20 minutes) in which I work through a developing taxonomy of these collaborations – including (e.g.), at present, “infectious” and “coincidental” collaborations – and the utility (or not?) of such distinctions. For me, it is not so much the taxonomic structure, as such, that is important, but rather that such a structure may allow an entity such as a consortium, or an individual campus, or even an entity on campus (i.e. the library) to identify by comparison the kind of digital community to which it belongs. This is to suggest a handful of observations that I will offer as provocations: that there is a link between the kinds of digital project that will grow out of a particular digital culture and the kind(s) of collaboration with which one identifies, that the former is in fact limited by the latter, and that understanding these limits – and how to push against them – gets us closer to thinking about a digital scholarship for which “the project” is not the basic unit but “the collaborative community.”
Towards Collaborative, Accessible Journal Publishing
Hal Hinderliter
Scholarly journals are intended to disseminate new research and theories for critical evaluation within the academic community. In most cases, however, these communication vehicles are biased towards the needs of a sighted audience, delivered as vertical rectangles of printed paper or frozen within a digital facsimile of the codex (the ubiquitous PDF). Too often, authors fail to consider the implications their chosen output format will have in determining audience access.
Is the PDF file truly the best option for academic publishing? Does its convenience for sighted authors and readers ultimately justify its shortcomings? Or do other considerations, including accessibility, metadata support, and machine discoverability indicate the use of alternative, non-page-centric formats?
This presentation will outline a new set of best practices for academic publishing workgroups by examining production techniques employed at at several journals that publish in EPUB format (an accessible form of electronic book). Collaboration within the EPUB publishing process will be demonstrated using Adobe’s InDesign and InCopy, resulting in a reflowable document suitable for consumption by a broad range of audiences.
By taking a closer look at who benefits from (or is excluded by) current academic publishing methods, this study hopes to better illuminate the exigencies served by accessible document design – a goal that could positively impact all forms of academic communication.
Interacting with In Mrs. Goldberg’s Kitchen
Adam Hochstetter
slides
This short presentation will provide an overview of the interactive tools that are a part of “Jewish Life in Interwar Lodz” website. It will briefly discuss the photographic procedures for capturing panoramas and 3d objects before introducing a tool for displaying and interacting with them.
A major portion of the website is based on an award winning museum exhibit in Lodz. The initial user encounter with the site is through an interactive panorama based on this exhibit. This panorama allows users to explore the physical exhibit, easily find more information about selected objects, interact with 3d representations of selected objects and integrate relevant audio. The panorama was created using hdr photography while the 3d objects were created using photogrammetry.
We needed a free open-source, viewer for these digital objects that was compatible with modern browsers (not flash) and could be integrated into Omeka. To do this we turned to X3DOM. X3DOM is “an open-source framework and runtime for 3d graphics on the web.” It can be integrated into the HTML DOM and supports most HTML events. Using JavaScript we were able to create complex interactions with the model of the physical exhibit. Integration with Omeka allows for the easy display of relevant metadata though it can be used in other contexts with minimal alteration.
Computer-Aided Rhetorical Analysis with Supercomputing
Suguru Ishizaki
I propose a short presentation of our collaborative project with Pittsburgh Super Computing Center.
In the past 15 years, we have developed an interactive text analysis environment/tool for rhetoricians to uncover common rhetorical strategies used in corpora of texts. Our tool, called DocuScope, is a dictionary-based tagging engine that utilizes a rhetorical dictionary, authored by David Kaufer at Carnegie Mellon, which consists of over 40 million linguistic patterns of English hierarchically classified into over 2,400 categories of rhetorical effects. (http://www.cmu.edu/hss/english/research/docuscope.html). DocuScope has been used for many research projects in the past.
The goal of our current project in collaboration with Pittsburgh Super Computing Center is to develop a tool that will allow us to analyze textual data that are significantly larger, and with much faster speed. In particular, we are currently interested in analyzing the use of languages on the social media as well as on the Internet in general. We are exploring ways in which this environment may allow researchers to ask research questions that have not been feasible. Such research questions include (but not limited to): How are certain rhetorical strategies distributed geographically and over time? To what extent and in what ways rhetorical strategies propagate in our society? How does the sentiment of the public expressed on the Internet change over time in relation to social events?
So far, we have designed the overall architecture of the system, and developed a simple working prototype that will allow us to carry out various experiments. In this presentation, we plan to present an overview of our project with a particular case study that involves the analysis of twitter messages about the 2016 election. We are currently implementing a system for this case study, and plan to complete a small working prototype before the conference.
Competitors or Corroborators?: A Comparative Digital Analysis of the Hugo and Nebula Awards for Best Novel
Chris D. Jimenez
Short presentation: This project examines two annual sci-fi literary awards—the Hugo and Nebula Awards for Best Novel—and their overlapping networks of authors and publishers. Both awards are given out for science-fiction novels 40,000-words or longer as voted on by similar demographics (attendees of the World Science Fiction Convention in the case of the Hugo and members of the Science Fiction and Fantasy Writers of America in the case of the Nebula), and both span roughly the same time period (1953-present for the Hugo and 1966-present for the Nebula). Given the overlaps between these awards with regard to their similar history, judging criteria, and membership, we might wonder how much overlap exists in their actual conferment of awards to novels. Do literary awards seek to carve out their own individual niches, or do they instead seek to corroborate each other, ensuring mutual prestige? What sorts of collaboration exist between them? To answer these questions, I use two DH methods. First, I draw on network mapping via Gephi to visualize the overlaps of authors and publishers between the awards. Second, I use text mining and topic modeling to compare word frequencies between their competing corpora. Ultimately, by comparing the awards together, this project seeks to better understand literary award culture in general as well as the communities that support it.
Mediating Native Languages in the Navajo Verb Generator Project
Nabil Kashyap
The Navajo Verb Generator Project (NVG) is a case study in leveraging digital environments to coordinate a range of individual, institutional and linguistic contexts. For Navajo, a consensus model of the language, particularly rules governing verb generation, is still a vital line of inquiry. NVG, led by professor Ted Fernhald, attempts to implement and test linguist Leonard Faltz’s influential model. Rather than an encapsulated set of scripts run through the command line, the project attempts to address multiple perspectives in the way it considers how a language might be represented in code. Navajo is distinct from other Athabaskan languages for many reasons. For one, it has the most speakers of any native language in North America. While members of the nation continue to learn Navajo actively, Navajo as deployed by younger speakers does not always involve fully formed verbs. At the same time, the language has had a particularly rich history of encounters with developing technologies, from World War II code talkers to early experiments in rendering Navajo on the web, and later, on mobile devices. On the front end, NVG is a bilingual web application that allows for public interaction and comment. The backend architecture is designed around the premise that undergraduate programmers with varying degrees of expertise will be able to modify the model over several years. In terms of the process itself, design and implementation has involved live feedback from a summer language institute based in the Navajo Nation, and in the summer of 2016, the project will include a remote team of students from Navajo Technical University in addition to those from Swarthmore. This presentation examines how NVG, both in its code and its development model, attempts to engage not only structural linguistic research but also the sociotechnical history of a dynamic indigenous language.
The Ethics of Networked Pedagogies: Examining the Collaborative Learning Practices of Online Gaming Communities
Matthew Kelly
In his article “Games, Learning, and Society,” Kurt Squire underscores the importance of examining the learning practices that occur in video games and gaming communities, stating that “today’s digital cultures are deeply participatory, using digital authoring tools and distribution networks to dramatically alter how we interact” with one another. Squire’s underlying argument is that the pedagogical practices surrounding video games are themselves reflections upon, and even interventions within, dominant cultural trends conditioned by contemporary economic, technological, and socio-political circumstances. For this paper, I plan to build upon Squire’s argument by examining the collaborative learning practices surrounding the acclaimed online role-playing game, Dark Souls. More specifically, I want to demonstrate how these communal pedagogical strategies reflect evolving notions of ethical responsibility within shared networked environments.
As I will argue, the collective educational resources created by the Dark Souls gaming community—including fan-made, machinima-style videos, online wikis, and asynchronous in-game messages—differ from traditional strategy guides that seek to endow the player with technical mastery of gameplay mechanics or help her overcome in-game obstacles. Rather, the learning practices and resources centered around Dark Souls attempt to incite a critical self reflection on both the ethical responsibility and personal vulnerability that players share towards each other. The ultimate aim of this presentation is to explore two interconnected questions: first, how can examining the learning practices which occur beyond traditional classroom environments grant insight into new ethical concerns that arise in networked societies and virtual spaces? Second, how can Digital Humanities scholars reflect upon these teaching practices as a way to better understand the ways in which our own research resonates with these evolving concerns regarding ethics in virtual spaces?
When Literature Refuses to Act its Age: Large-Scale Stylochronometry among the Weird Genres
Matthew Lavin
Stylochronometry is defined as the use of statistical and often computational methods to analyze the date of inception of a text or group of texts. Divided into two major branches—a relative and absolute dating—the field of stylochronometry has much to offer digital humanities. As scholars such as R.S. Forsyth and Constantina Stamou have demonstrated, “relative dating” applications can provide insights into how an author’s style changed over time, or which manuscript of a particular text is the oldest version. Ted Underwood and Jordan Sellers have used techniques closer to “absolute dating” to trace the divergence of “literary diction” from other published work between 1700 and 1900.
In my individual conference paper for Keystone DH, I will focus on the application of stylochronometry to a corpus of “weird fiction,” a collective term often used to describe gothic, horror, and otherwise supernaturally-themed genre fiction. This Spring, using resources of the Pittsburgh Supercomputing Center, I am collaborating with two partners, Alexander Gladwin and Daniel M. Look, on a project that examines H.P. Lovecraft’s use of archaic terms and neologisms. The case of HP Lovecraft is an intriguing one because he made various attempts, many quite aggressive, to alter his literary voice to sound archaic. However, in my Keystone DH presentation, my question will be to what degree that impulse can be seen in the weird genres more generally. I will consider texts before and after Lovecraft’s era, as it will be interesting to see if he was participating in or possibly spurring on a larger trend, but my goal is not merely to advance to the field of individual author studies around Lovecraft and his work. Rather, I will address what stylochronometry can tell us about how the diction and style helped authors of weird fiction position their work as belonging to older periods.
Interdisciplinary Approaches to Metadata
Tom Lombardi
Presentation Slides
Modern research has produced a stunning amount of data in virtually every field of study. For example, biologists in the age of gene technologies and bioinformatics have had to grapple with this volume of data for about two decades. In particular, new approaches to evaluating metadata have been developed to address the growing need for the analysis of such data. This paper explores the possibility of applying techniques developed for analyzing metadata in disciplines like biology to comparable metadata in art history. In particular, the research outlines the application of such techniques to the wealth of data stored in the Index of Christian Art (https://ica.princeton.edu/). The preliminary research aggregated data for over 1500 Tuscan paintings from the Fourteenth Century to identify subjects handled differently before and after the Black Death. The results demonstrate shifts in the iconography of Mary consistent with scholarly interpretations. Despite some interpretive challenges, the early results of this work suggest that such techniques could be repurposed to provide advanced search features to large metadata collections in art history. The comparison of interdisciplinary approaches to metadata offers digital humanists the opportunity to reflect on data-sharing practices in the discipline and explore the rich set of options for organizing the metadata supporting research.
Emily Dickinson: Fascicle 16
Nicole Lottig, Brooke Stewart, and Brooke Lawrence
Emily Dickinson: Fascicle 16 is a repurposed version of two previous websites that explore the poems in Dickinson’s Fascicle 16. The original websites look at variants within the poems that Dickinson herself had made. Our group compared the changes that were made to the poems in a selected list of published versions from the nineteenth and twentieth centuries. We worked together in our Fall 2015 DH class at Pitt-Greensburg to completely refurbish the existing work and display our own findings.
We worked with photo-facsimiles of Dickinson’s manuscripts to fix any transcription errors and mark-up the poems’ original variations using TEI XML coding, and we added the variations found in the published versions. Next, we created a complicated XSLT to transform our XML into HTML. Using CSS and JavaScript we were able to display the variations between the manuscript transcriptions and the published versions and project a visually pleasing website. In order to show where the variations were located and compare the manuscripts to the published versions of the poems, we utilized Photoshop and image-mapping over all of the poems’ images. Finally, we produced an SVG graph to represent the percentage of dash reduction in the published versions compared to the manuscripts.
Our studies found that the published versions changed the original poems considerably. We particularly focused on the dashes that were removed by later editors. These changes are significant because the dashes sometimes create a caesura, and the lack of or addition of them in the published versions alter how the poems are read and understood. This knowledge will help facilitate other scholars’ literary research about Emily Dickinson’s Fascicle 16. Future development of this project will include analyzing other changes between the manuscripts and published versions of the poems, such as capitalization, punctuation, and word choice.
Augmented Reality Collaboration: The Augmented Palimpsest Project
Tamara O’Callaghan and Andrea R. Harbin
Presentation Slides
Our digital humanities project, The Augmented Palimpsest explores how the medium of Augmented Reality (AR) can be applied to teach medieval literature effectively. Using Chaucer’s General Prologue, the tool delivers digital enhancements that emerge from the printed page via a smart device. We are creating simple print pages with highly detailed manuscript borders set around the text of Chaucer poem. As such, each page has the appearance of a medieval manuscript folio with a border that is coded with a variety of digital enhancements, including but not limited to audio, video, and graphical materials; and 3D models of figures, architecture, and objects. The student opens our AR application or “app” on a smart device, such as an iPhone, iPad, or Android tablet/phone, and then holds that device over individual fiducials embedded in the border to access the various enhancements coded to each fiducial.
Thanks to an NEH Digital Humanities Start-Up Grant, we are now building the tool with a team comprised of a computer scientist and AR expert, two literature professors at different institutions, and a couple of graphic artists. This paper will introduce the project and discuss how collaboration across disciplines has made this work possible as well as exploring some of the pitfalls we have encountered working with colleagues who are not only working within other disciplines, but who are also removed from us geographically. This project demonstrates that such collaborations are not restricted to large research institutions, but that with collaboration and planning may be accomplished even from within smaller institutions with more limited resources.
Topic Modeling Communities of Discourse in Doctoral Dissertations
Benjamin Miller
In this 20-minute presentation, I will introduce and invite discussion of interactive data visualizations derived from a computational content analysis of roughly 2,000 full-text doctoral dissertations in Composition/Rhetoric. Because this corpus is larger than any individual researcher could read, the analysis relies on on topic modeling: an algorithmic means of identifying clusters of words which tend to co-occur within documents, and the proportions with which these word clusters contribute to the documents (cf. Blei et al; Mimno; Goldstone and Underwood; Ratliff). Interpreting such topics and proportions in light of the dissertations’ abstracts, I offer a concept map of subject matter and key terms in recent circulation among young scholars entering the field.
In addition to identifying the most frequently occurring topics in the corpus (including pedagogical theory, philosophy of language, and identity construction), I will explore multi-topic clusters emerging from visualizations of topic-topic similarity. Such clusters both complicate and support impressions of subject matter derived from the frequency of individual topics alone; they also help focus further study into subtle distinctions of terminology used in discussing related subjects. In the case of this corpus, data visualizations based on topic co-occurrence within dissertations suggest that combinations across topic clusters are common, but not evenly distributed. “Missing” combinations can then suggest new possibilities for future dissertation research, making this kind of visualization not only an analytical tool, but a generative one.
Using IIIF for Small Projects
David Newbury
Presentation Slides
IIIF, or the International Image Interoperability Framework, is an emerging Linked Open Data standard for image interoperability. It defines metadata standards for dealing with high-resolution images, providing a consistent API for accessing both images, the metadata that surrounds them, and how to present and associate images together. It is being used at the Internet Archive as well as major museums and national libraries around the world.
By employing this emerging digital standard to host image metadata, it allows image resources to be easily shared, incorporated, and recontextualized without loss of authority or human intervention.
While the standard is comprehensive and extremetly useful, often the infrastructure requirements to deploy IIIF appear to be out of the scope of smaller projects and institutions. As part of the new archival website at the Carnegie Museum of Art, we have identified techniques and developed open source tools that allow institutions and projects to implement the base profile of IIIF on a shoestring budget, using Amazon S3, spreadsheets, and other simple tools.
I propose a short presentation providing an overview of IIIF, a demonstration of its use at other institutions, a review of how CMOA is using this tool to facilitate sharing of images, and an brief explanation of how other institutions can use our tools to facilitate sharing their images using IIIF.
Authority Cascades for Linked Open Data
David Newbury
Presentation Slides
As Linked Open Data is increasingly embraced throughout the cultural heritage sector, we are moving from an environment of data scarcity to an environment of data overload. For example, often we no longer have a single authority for the names and identities of the participants in some activity or event, but two, three, or even more. Getty ULAN, VIAF, Wikidata, and other authorities provide a wealth of information about important people throughout history. However, no authority is comprehensive, and due to the Open World Assumption different authorities may provide conflicting or inconsistent information about a single participant.
While Linked Open Data is designed to accomodate these conflicts at the data level, this new profusion of authority records presents problems and requires new strategies for dealing with overlaps and conflicts in the presentation of Linked Open Data. Also, often there exist records that do not exist in any authority, but that emerge through research and must be incorporated in any given project.
I propose a short presentation on authority cascades, a practical technique for both querying and accomodating these inconsistencies in a logical, easily understandable way. This technique allows for institutions or projects to leverage Linked Open Data, while remaining their own “authority of last resort”. By cascading down a implementation-defined list of authority sources, a single, comprehensive presentation of information can be created.
Git Started!
Rebecca Parker
Presentation Slides
Managing a project team especially of undergraduates requires organization, accountability, and lots of GitHub activity. This presentation will take a behind the scenes look at the GitHub collaboration of Pitt-Greensburg’s student projects and the DHClass-Hub. Rebecca Parker, the 2015-2016 teaching assistant to Dr. Beshero-Bondar of the University of Pittsburgh at Greensburg, will detail the important uses of Git’s version control system and the social media aspects of GitHub that make it an ideal set of tools for project collaboration. She will demonstrate her personal use of GitHub as well as the unique uses of GitHub that are experienced by Pitt-Greensburg students. An overview of how to use Git shell (the command line Git interface) will encourage the audience to “look under the hood” of the Pitt-Greensburg Digital Humanities Courses. Audience members can expect to leave this presentation with the ability to collaborate on projects of any size using Git. A run through of Rebecca’s CitySlaveGirls GitHub Repository will model how a project with a fluctuating team of editors can maintain project stability with GitHub’s detailed repository history recording and collaborator tracking. The main goal of this presentation will be to orient the audience with Git and GitHub in hopes of changing the stigma of confusion and difficulty surrounding their use.
Building and Managing a Digital English Studio: Collaborative Aspects
Stuart Selber, Daniel Tripp, and Leslie Mateer
Presentation Slides
Digital Humanities projects vary widely in their requirements for institutional support. Some people are able to work quite independently with as little as a laptop, public WIFI, and an open-source application or two. Others, and we would argue many others, are working on projects that require significant resources, social and technical. Academic institutions tend to be good at providing support for a general university population, but that type of support is often inadequate in specific situations. How might departments and programs support the individual digital projects of humanities scholars and teachers? And what types of collaborative structures are helpful to the endeavor of creating supportive environments for specific purposes?
We propose to discuss how we built and manage a Digital English Studio at Penn State, especially the collaborative aspects. The project portfolio of the Studio is diverse and expansive: We create and support the creation of websites, personal and departmental; we manage the department’s social media assets and supervise undergraduate social media interns; we design and oversee the department’s distance education courses; we support teachers who are working to earn a Teaching with Technology Certificate; and more. The staff of the studio includes a Director, Managing Director, and four Studio Assistants (two fixed-term faculty members and two graduate students). The hardware in the Studio includes iMac workstations, a smart TV, a scanning station, and a video and audio recording room. We also have equipment people can borrow, such as iPads and digital cameras. The Studio is staffed for 10-12 hours a week, but we can also give users around-the-clock access via a university ID card swipe.
Our discussion will focus on a number of collaborative relationships that have been essential to the enterprise of building and managing a digital studio: Our relationship with our funding sources (the College of Liberal Arts and the English department); our relationship with our clients (humanities scholars and teachers); and our relationship with campus computing services, which are necessarily entangled with our own technical and social infrastructures. We will characterize the nature of these collaborative relationships and outline strategies for creating productive campus partnerships that can support individual digital projects in the humanities.
On the Diversity of Digital Decay
Robert Sieczkiewicz
The problem of digital decay is well-known, from link rot in Supreme Court opinions to prize-winning journalism that disappears from the internet when a newspaper folds. In recent years, there has been some study of the decay of early digital humanities projects. What has not been addressed, however, is whether digital decay affects DH projects unequally. This study hypothesizes that DH work on black writers and black history is more vulnerable to digital decay than other DH work.
In their 2009 survey on “graceful degradation” of digital humanities projects, Bethany Nowviskie and Dot Porter reported that 64% of survey respondents experienced decline or difficult transitions in their projects; of those, 26% of the projects were “abandoned or dormant.” Robin Camille Davis found that in 2015 “just over half of the 60 projects presented at DH2005” were still online. This study will build on this baseline understanding of widespread DH decay by counting sites that were operational in 2005-2006 and determining how many were defunct in 2016. Then study will identify sites from 2005-2006 related to African-American, African and Afro-Caribbean writers as well as sites about African-American, African and Caribbean history, and compare their rate of digital decay to the larger set. If a difference in decay rate is confirmed, the study will seek to identify the reasons why.
Preliminary research results will be presented at KeystoneDH16, and will be used to consider whether the DH community should modify its collaborative preservation efforts to ensure the survival of DH scholarship on black writers and black history.
Representing the Un/der-represented: Using Data Visualization to Explore Diversity and Inclusion for Academic Library Collections
Bobby L. Smiley
Without institutional knowledge, or recent collections analysis, it is difficult to get purchase and perspective on how large, research intensive academic libraries address diversity and inclusion in their existing collections. Using Michigan State University Libraries’ collection of history holdings as an example, this presentation examines how data visualization furnishes pictorial, quantitative ways to discern trends, identify gaps and silences, and assess collection strengths in diversity and inclusion to help inform future collection development decisions.
Building on scholarship addressing diversity in collection assessment, this presentation unfolds in two parts: first, it briefly contextualizes diversity and inclusion in library collections (and concomitant problems of relying on Library of Congress Subject Headings for analysis). The balance of the presentation is dedicated to exploring the application of data visualization tools and techniques to library data gleaned from the Statistical Category (SCAT) table of Michigan State’s collection, circulation statistics, and use-ratios. Visualizing this data (using treemaps, bubble clouds, network analyses, among others), the presentation focuses on how data visualization can reveal with immediacy collection size proportion, where silences and strengths reside, and use by LC class. In particular, Michigan State’s collection of history and history-related materials (Library of Congress Classification C-F) provides an example of how visual and quantitative data enable easier, opportunities for critical re-evaluation of a collection, as well as evidence for acknowledging and redressing gaps in collection diversity and inclusivity. The presentation concludes by leveraging information to consider how libraries can use data visualization (and other digital humanities approaches) to build that are truly representative and inclusive.
Exploring the Adoption of STEM Praxis in the Digital Humanities
Sarah C. Stanley and Matthew E. Hunter
Presentation Slides
As digital humanists assert themselves into the academic milieu, digital projects increasingly leverage scientific praxis for humanistic inquiry. The adoption of these practices engenders seemingly-ubiquitous digital scholarship buzzwords such as “project management,” “collaboration,” “interdisciplinarity,” and (in the scariest scare-quotes of all) “funding.” These concepts appear to be at odds with the classic image of the lone humanities scholar. While STEM researchers are accustomed to considering the influence of funding agendas on scholarship, collaborative attribution models, and product-driven scholarship, these are emergent questions for humanists.
We interrogate W. H. K. Chun’s provocation that the digital humanities confront binary “light” and “dark” sides in adopting STEM practices, and offer that the “fears and biases [in creating] deeper collaborations with the sciences and engineering” are imperative to analyze; that “forging joint (frictional and sometimes fractious) coalitions“ outside the humanities is vital to the discipline as it causes humanists to question their own scholarly doctrines. We propose a dialogue that will explore both the impact of adopting STEM models in the humanities and the potential for collaboration between humanists and STEM scholars. We will specifically focus on how scientific collaboration models might affect collegiality and attribution in DH scholarship and how resistance to finality in humanities scholarship can function in a deliverable-driven funding landscape.
This dialogue is informed by the authors’ academic perspectives as humanities-trained librarians and scholars, and as professionals within the University system. Matthew facilitates predominantly STEM research as a University grants officer, and therefore understands the financial and administrative intricacies of the funding landscape. Sarah works as digital scholarship support staff in the university library, and has firsthand experience in digital humanities research. These backgrounds inform our discussion on the challenges and benefits of collaborative and interdisciplinary methods of humanistic scholarship.
Tracing the Influence of “The Right to Privacy”
Susan Tanner
This works-in-progress presentation will deliver preliminary findings from a corpus analysis of several key legal texts and privacy law decisions. Numerous legal scholars have theorized that “The Right to Privacy” is one of the most influential law review articles ever written (see, e.g. Posner; Solove; Cate; Gormley). My research is designed to test that theory. I trace evidence of textual influence from “The Right to Privacy” through contemporary legal and non-legal discourses about privacy rights. To do so, I employ techniques of micro analyses of intertexuality on a much larger scale.
My presentation will focus on the first part of a two-part project wherein I study both legal and popular discourses. The first stage of my project has been to collect a corpus of legal texts dating from circa 1850 to the present and classify them into genre categories. I have identified key variations of Warren and Brandeis’ definitions of privacy in order to construct a timeline of legal influence. Part of the purpose in reconstructing a line of reasoning for privacy law decisions beginning with Warren and Brandeis is to fully map one line of precedent from its inception to the current day through its iterations in Supreme Court opinions. This serves the dual functions of examining one line of reasoning brought to bear on contemporary legal analysis and formulating a corpus of arguments that will provide data from which I will make claims as to what comprises standard appellate discourse. From the corpus, I am building dictionaries of words and phrases that appear and reappear as legal and rhetorical concepts important to an understanding of privacy as it is understood by the courts.
Translation Analysis with TEI: Robert Southey’s Amadis of Gaul
Stacey Triplette and Elisa Beshero-Bondar
Presentation Slides
The late fifteenth-century Iberian romance Amadís de Gaula offers an attractive opportunity for translators because it is unmoored in time, space, and authorship. Medieval versions of the story have been lost, and sixteenth-century editions of the text often stripped out the name and personality of Garci-Rodríguez de Montalvo, whose edited Amadís dates to 1508 and forms the basis for modern critical editions. Though twentieth-century discoveries reveal a Castilian origin for the work, in the sixteenth through nineteenth centuries it was often thought to descend from a French or Portuguese original. Southey, translating the work into English in 1803, sustained the Portuguese thesis, and his Amadis compresses and alters the original text according to the aesthetic criteria of the Romantic era and the nascent British Empire while purporting to restore the simplicity of an idealized Portuguese ur-text.
Our digital project uses TEI to align Southey’s text with the Spanish “original” he used, a 1547 Sevilla printing. This presentation shows how the TEI markup language can reveal translation choices by allowing us to track and quantify omissions, additions, and rewritings of the Spanish text. Our project reveals the correspondences and contrasts between Southey’s translation theory, articulated in his Preface to Amadis, and his practice, allowing us to form new hypotheses about how translation works at the level of the clause and the sentence. Our project deploys the TEI in some unusual ways to help us align parallel passages and consider what was reduced and superadded in the translation process. We will discuss our markup, data extraction, and some experimental outputs in charts and SVG graphs, and our efforts to test and formalize a methodology for fine-grained analysis of translations. We will share our digital methods for interdisciplinary collaborative research on translation as a work in progress.
Tag-Team Teaching a DH-Inflected Diversity Course
Linda Troost and Charles Hannon
Five faculty members from four departments (Computer and Information Science, English, Russian, and Spanish) created and taught “Identity, Ethnicity, and the Digital Humanities,” a course for undergraduates in which technological tools were used to investigate representations of identity and ethnicity in literary and cultural texts. The course also satisfied student-learning outcomes in two general-education areas (Humanities and Diversity) and one program (English, where our course was housed).
In Fall 2015, ten undergraduates learned to grapple with complex issues of identity and ethnicity; hone their literary skills of close reading, genre analysis, and contextualization; and develop their academic skills of essay writing, oral presentation, and project management. In addition, they learned new digital methods to aid humanistic inquiry (database mining, computerized text analysis, mapping, data visualization) and new digital tools to present their findings (blogging, timeline, and video-editing apps). The content of our course ranged widely, including historic and current material, many genres (films, drama, fiction, newspaper articles, maps, memoirs), and several geographical areas (the U.S., Latin America, England, Ukraine). We analyzed African American heavyweight boxers, Latino culture, the English slave trade, native languages in Crimea, and family relationships in the American South. Final student projects built upon these subjects.
In our talk, two of us will discuss the course, with particular attention to (a) strategies for teaching the beginner in textual study, diversity issues, and digital method, (b) balancing the teaching of subject matter (e.g., language and identity) with the teaching of skills (e.g., learning Tableau) and, (c) collaborating across departments to address workload and organizational issues.
Allá y Aquí: American and Mexican News Perceptions of Ciudad Juarez
Roberto Vargas
If you know of Ciudad Juarez, Mexico, then you know of its claim to fame: violence. This one-dimensional view of Juarez is reflected in multiple ways through many outlets: academic papers, books and, more widely, the news. In this short presentation (7 minutes), we will demonstrate the early progress and development of “Allá y Aquí: American and Mexican News Perceptions of Ciudad Juarez ,” an ongoing project that seeks to compare and contrast how the news outlets represent and write about Ciudad Juarez here in the United States versus in Ciudad Juarez. Using known tools for text analysis, we seek to test whether geographical distance affects how writers think of the city and whether the missing vocabulary on both sides can tell a story that is different than the one written. As the project continues to develop and expand over this upcoming year, in collaboration with the Swarthmore College faculty, we will try to deal with questions of English and Spanish pragmatics (from a general linguistics sense) and to explore the textual impact context has on what is written. Knowing the ambitious nature of this project, we understand that there are many questions we have not asked and will continue to refine our approach as we feel deeply connected to this city and how others perceive it.
Gender and Centrality in the Digital Humanities
Scott B. Weingart, Nickoal Eichmann, Jeana Jorgensen
Presentation Slides
The annual international Digital Humanities conference is famous for its rhetoric of inclusion, with some recent conference themes including “Big Tent DH”, “Cultural Empowerment”, and “Global DH”. Recently, feminism and gender studies have been particularly audible topics of discussion surrounding these conferences, and many of the most recognizable voices in the community are women’s. In 2015, nearly half of all conference attendees were women.
This is difficult to reconcile with the fact that, since 2000, women made up only a third of named conference presentation authors, and in those presentations the topics of gender and feminism are rarely discussed. Indeed, those topics and related ones are systematically biased against during the conference’s peer review phase. These findings are part of a study exploring 15 years of DH conference proceedings, the results of which will be presented here, along with a discussion of what is driving this disconnect and how it may be resolved.
Analyzing Ethos: Developing Digital Tools for Argument Analysis in the Humanities
James Wynn and Rick Costa
Rhetorical training in argumentation has been a central part of humanist education for over two thousand years. As part of this training, students of the humanities learned argument could be developed using appeals to good reasons and evidence (logos), emotion (pathos), or the character of the speaker (ethos). In the modern era, digital technologies provide an unprecedented opportunity for studying these features of argument and comparing trends in their use on scales unimaginable to previous scholars of rhetoric and argument. Despite these obvious benefits, significant obstacles face digital humanists interested in studying textual phenomena of argument that go beyond the word or short phrase. In my talk I will discuss my current project, e-thos, an analytical platform designed to identify appeals to expert ethos. My presentation will demonstrate the range and limits of e-thos by examining its capacity to identify expert appeals in scientific Congressional testimony on climate change. It will also discuss the contributions that the tool makes to the argument theory of ethos, to our understanding of scientific argument over climate change, and to the development of digital analysis platforms for humanist scholars.
What Might an Archive “Know”?: Annotation through Recursion in Digital Mitford: The Mary Russell Mitford Archive
Mary Erica Zimmer, Molly O’Donnell, and Elisa Beshero-Bondar
Presentation Slides
In a keynote to the University of Michigan’s 2016 symposium on “Text Mining Across the Disciplines,” Laura Mandell argued for oscillating between qualitative and quantitative methods, as a means of placing corpus-based insights in context. Such a call likewise suggests large-scale archival approaches as able to inform more discrete editorial questions. As Elena Pierazzo (2015) observes, contemporary editorial projects ask scholars to serve as part of diverse teams in new and emerging environments. This paper will explore how existing approaches to scholarly annotation may yield models for computationally informed praxis.
Specifically, the editors of Digital Mitford will report on work in progress allowing an archive’s own texts to bring forward the scholarly, expert context annotations require. Our approach involves mobilizing insights arising from our team’s own TEI markup. Using XQuery on our database of encoded documents, our Annotation Tool assesses the prevalence of relational categories tagged via our editorial principles, then uses these counts to weight lists of high-frequency tokens in paragraphs indexed by a key term. The resulting clusters of knowledge suggest lines of qualitative inquiry, while quantitatively enriching the team’s perspective on entities of interest.
Centrally, such archival clusters themselves also serve as forms of annotation—ones that, rendered judiciously, aid the work of editors and readers. To show our approach’s potential, we will focus on the challenge of locating one mysterious figure, “Miss James,” about whom discrete editorial conjecture began in Spring 2015, when several members of our team noted her presence in letters from 1819 and beyond. While these glimpses suggested the charismatic “Miss James” as Mary Mitford’s trusted friend and advisor, conventional means of tracing this individual’s biography proved elusive. By mobilizing the potential insights a TEI-tagged corpus may reveal, our approach is allowing both editors and readers to map more precisely her place in Mitford’s world.
Panel: Videogame Adaptation
Bradley J. Fest, Kevin M. Flanagan, and Jedd Hakimi
Hakimi Presentation Slides
As videogames continue to emerge as a dominant twenty-first-century form, it is becoming clearer that they have complex relationships to other media. This panel, part of a larger collaborative project, will address issues of adaptation and videogames from a transmedia perspective, drawing particularly on the resources provided by film and literary studies.
“Videogame Adaptation: Some Experiments in Method” - Kevin M. Flanagan, University of Pittsburgh
This paper outlines the concerns and conceptual practices of videogame adaptation, noting the many ways in which videogames shape, or are shaped by, ideas, narratives, and mechanics from other media. In situating videogames into the discourses of textual transformation that animate current work in adaptation studies, I argue that traditional approaches to adaptation in English departments (which privilege novel-to-film adaptation in a one-to-one correspondence) have a lot to learn from games, which function as adaptations at all stages of their production and consumption. I also demonstrate how adaptation studies challenges claims to medium specificity that form a foundational conceit of videogame studies.
“Metaproceduralism: The Stanley Parable and Postmodern Metafiction” - Bradley J. Fest, University of Pittsburgh
Most critics of contemporary literature have reached a consensus that what was once called “postmodernism” is over and that its signature modes—metafiction and irony—are on the wane. This is not the case, however, with videogames. In recent years, a number of self-reflexive games have appeared, exemplified by Davey Wreden’s The Stanley Parable (2013): an ironic game about games. When self-awareness migrates form print to screen, however, something happens. If metafiction can be characterized by how it draws attention to language and representation, this paper will argue how self-reflexivity in videogames is best understood in terms of action and procedure, as metaproceduralism.
“Playing Los Angeles Itself: Experiencing the Digital Documentary Environment in LA Noire” - Jedd Hakimi, University of Pittsburgh
Almost everything about the predominantly faithful depiction of 1947 Los Angeles in the recent, police-procedural videogame LA Noire (2011) was based on archival material, including period maps, photography, and film footage. And while scholars have thought extensively about how film spectators experience mediated depictions of real-world cities, the videogame player’s parallel experience has been relatively unexplored. Accordingly, I take LA Noire’s simulacrum as an opportunity to reflect on what happens when a real-world environment is adapted into the setting for a videogame. Specifically, I position LA Noire in the tradition of the “city-symphony” film and a particular sub-set of Film Noir known as the “semi-documentary” to make the case LA Noire contains crucial aspects of the documentary image. Consequently, LA Noire is not so much creating a fictional, diegetic world, as it is presenting our own world back to us in a manner that changes the way we experience the world in which we live
Workshops
TEI for Historical Manuscripts and Letters: A Beginner’s Workshop on Transcription, Metadata, Paleography and Code
Elisa Beshero-Bondar, Lisa Wilson, and Amy Gates
Presentation Slides
Join members of the Text Encoding Initiative (TEI) and editors of the Digital Mitford project to gain hands-on experience with transcription and coding of manuscript letters. We will work with a selection of photo facsimiles and partially encoded transcriptions of letters by Mary Russell Mitford, as well as with the TEI Guidelines and
Participation in this workshop will require participants to install a (free) version of the
Workshop Materials - Participants are not required to download and install oXygen, but we invite them to do it if they’re interested.
Curating the Digital: Collaboration and Transformation
Stefanie Dennis Hunker, Jolie Sheffer, and Carol Singer
With the availability of digital curation tools and digital environments, we have the opportunity to think about a new era in the study of literary, textual, and material culture. We are in a new world, where visual images circulate, are contextualized, transformed, and shared across the globe, in an instant. Yet now more than ever, these images require context, historicization, analysis, and interpretation. The use of these digital tools can actually transform how we study culture, and how we communicate our findings with others. Thus, digital curation has profound implications for pedagogical practice, for processes of cultural analysis, and for the audience(s) for library archives.
This workshop session is focused on pedagogical considerations and strategies for using archival materials as the basis for curated digital exhibits. The session is designed for archivists, library staff, and humanities faculty interested in having students work with digitized primary source materials. Based on several years of collaboration between archivists and instructors at Bowling Green State University, this session’s facilitators will share best practices for effective collaborations. We begin by discussing the goals of such projects, and identify a variety of tools that can be used to facilitate digital curation projects, including PowerPoint, Tumblr, Omeka, and WordPress. We will then discuss key considerations in developing institution-specific (and course-specific) processes for curating archival materials, such as audience, copyright and reproduction restrictions, workflow, metadata, time management, and technology. The majority of the workshop will be interactive, with participants curating freely available digital items for metadata, and creating exhibits on OMeka.net. We will guide participants in developing a tailored process, from identifying faculty partners to engage students in digital curation projects, to managing workflow, utilizing various technology tools, and recontextualizing materials to create different narratives.
Project Showcases
Ticha: The Story of an International, Community-Engaged Digital Humanities Project
Laurie Allen, Brook Danielle Lillehaugen, George Aaron Broadwell, Mike Zarafoneti, and Michel R. Oudijk
Presentation Slides
The project demonstration will introduce Ticha: A digital text explorer for Colonial Zapotec, and will focus on the collaboration between linguists, ethnohistorians, librarians, undergraduates, and modern Zapotec language speakers. While still relatively young, the project has already resulted in the Ticha website which allows users to view and explore multiple layers of 16th to 18th century texts written in Zapotec, an endangered language, indigenous to Oaxaca, Mexico. The website brings together both pre-existing digital images of printed texts from US Collections, including the John Carter Brown library, and new high resolution images of manuscripts handwritten in Zapotec, such as wills and bills of sale, the originals of which are stored in diverse archives in Mexico, sometimes difficult to access. Each text is made available publicly on the Ticha site throughout the process of transcription, encoding in TEI, translation, normalization, and additional annotation, including linguistic analysis. Modern Zapotec speakers have been consulted throughout the project, and their input has helped shape both the current website and priorities for the future. During this presentation, we will talk about the collaborative process of creating this interdisciplinary project, including how the team has worked together across international, professional, and language boundaries to develop a site that will be useful for multiple scholarly and public audiences.
Text Annotation Modules and 19th-Century Literature
Todd Bryant & Sarah Kersh
Presentation Slides
During the summer of 2015 the Mellon Foundation Digital Humanities grant at Dickinson College funded a project to produce an online, annotated edition of a volume of poems written by the 19th-century poet Michael Field, https://michaelfield.dickinson.edu/. The team, led by Sarah Kersh, Assistant Professor of English, and Todd Bryant, Language Technology Specialist, consisted of two undergraduate English majors, and one student from computer science.
Michael Field’s Sight and Song (1892) is a volume of ekphrastic poems, (i.e. poems written about paintings), and our annotated edition allows viewers to see the poems alongside their associated images. Annotations within the poems provide background information making Sight and Song more accessible for students and teachers alike and the project brings to the forefront the current relevance of a text previously marginalized and currently out of print.
The site is built on Drupal and utilizes other open source tools and modules of use to the wider digital humanities communities. The annotations are created using a custom Drupal module developed by Sai Grandhi which integrates with an open JavaScript library AnnotatorJS, the same library used by Hypothesis https://hypothes.is/ and Annotation Studio http://www.annotationstudio.org/.
The visualizations created by English majors Georgia Christman and Katie Jarman using Gephi show models of repeated patterns and key words present in the poems and the paintings. Going forward, we plan to use data from the Hathi Trust to look at word frequency across time to better situate Michael Field’s work within the sphere of nineteenth-century texts.
Todd Bryant will be happy to show anyone interested how to implement the AnnotatorJS JavaScript on Drupal or elsewhere, use Gephi, and how python scripts can be used to combine and modify the data sets from the Hathi Trust.
Broken Books
Debra T. Cashion
Presentation Slides
For many antiquarian books and manuscripts, the inevitable consequence of increased value over time results in an unfortunate fate: they are taken apart, “broken” into pieces, sold for profit among various owners and institutional collections. As scattered fragments, however, detached leaves present to scholars frustrating glimpses of historically significant texts and decorative programs that cannot be fully studied without evidence of the original work and the context in which it was produced. The Broken Books resource allows for the virtual restoration of these dismembered books, so that they can be digitally recaptured, reordered, and reconstructed, at least as much as preserved leaves of are found. Under development at the Center of Digital Humanities at Saint Louis University, this project uses IIIF technology and the Mirador viewer to create a web-based application and linked data environment that not only allows a user/administrator to manage a virtual reconstruction, but also permits other users to contribute images and information, including cataloging metadata, about undiscovered leaves.
The test case manuscript for the project is the Llangattock Breviary, a 15th-century Italian illuminated manuscript that was broken apart by an American bookseller after it was sold at auction in 1958. There are leaves from this manuscript in collections and institutional repositories all over the world, including Harvard, U.C. Berkeley, the American Academy in Rome, University of South Carolina, Michigan State, Dartmouth, and the Hill Monastic Manuscript Library, Washington University in Seattle, the Louvre Museum, and the Museo Schifanoia in Ferrara. Thus far a total of 67 leaves have been virtually united through the Broken Books digital resource, a demo version of which can be viewed online at: http://165.134.241.141/brokenBooks/home.html?demo=1. For further information, see my Omeka website: http://brokenbooks.omeka.net/exhibits/show/llangattock.
A Digital and Naturalistic Landscape of Thomas Hardy’s Wessex
Erica Y. Hayes
Presentation Slides; Project Website
This project showcase exhibits the creation of a GIS story map of Wessex, the fictional setting in the novels and poems of nineteenth century British author and poet, Thomas Hardy. Based on the six counties in the southern region of England, Hardy named his imaginary landscape after the medieval Anglo-Saxon kingdom that existed there prior to the Norman conquest. Wessex first appeared in chapter 50 of the serialized version of Hardy’s novel, Far from the Madding Crowd. Due to the positive responses Wessex received from readers, Hardy continued using the fictional landscape in his next novels. In his sixth novel, The Return of the Native, he began to divide Wessex into different regions. When it came time for Hardy to revise his earlier works for his newer editions, he entitled his revised editions as The Wessex Novels with an illustrated map of the imaginary landscape. For this reason, when we read Hardy’s novels today, we get a fully-realized picture of Wessex in all of his novels. Due to the central element Wessex serves in Hardy’s fiction, mapping this imaginary landscape is highly ideal. Using geospatial humanities, this project allows scholars and students to visualize the towns set in Thomas Hardy’s Wessex and compare them to their real locations on a basemap of today’s southern region of England. Using the New York Public Library’s open source mapping tool, Map Warper, I have geo-rectified a digitized edition of Thomas Hardy’s illustrated map of Wessex and aligned it in ArcGISonline with a basemap of its corresponding real locations in today’s southern region of England. Starting with Hardy’s most famous novel, Tess of the D’Urbervilles, this digital map plots the novel’s narrative through its locations in the order in which they first appear in the novel. Considered one of the best examples of naturalistic fiction, this digital map explores Wessex through the natural world Hardy captures in Tess of the D’Urbervilles. More than a mere setting in his writings, the map allows readers and scholars to investigate Hardy’s fictional setting via locations annotated with passages highlighting the use of naturalism in the novel. This project was completed in collaboration with my DH classmates, the GIS Librarian at IUB Libraries, and the IUB Digital Humanities Cyberinfrastructure Coordinator. Future goals of this project include building upon this digital story map to include Hardy’s other Wessex novels and performing textual analysis of concatenated words in connection to the towns and places in Hardy’s novels—I would love to gain feedback about this project at Keystone to help guide the future goals of this project. As the study of geospatial humanities continues to grow and explores the intersection of space and literary scholarship, this digital humanities project hopes to offer new literary insights into Hardy’s fictional world, life and writings.
Exploring Place in the French of Italy
Heather Hill
Presentation Slides
From approximately 1250 to 1500, it became popular for writers in the Italian peninsula to create and copy works in the French language. Scholars studying these French of Italy texts have traditionally used close reading to explore the corpus, assuming a geographic orientation toward France. Newer scholarship challenges this notion, arguing that these texts taken together focus instead toward the east.
Fordham University’s Center for Medieval Studies’ digital project, “Exploring Place in the French of Italy” (EPFOI), supports the idea that French of Italy texts focus geographically on more than just France. Using a method that may be described as “mid-range” reading rather than close or distant reading, EPFOI uses geospatial visualizations to display identifiable places from a sample of sixteen texts from the corpus and reveal trends that not only indicate an eastern gaze in literature but also differences between conceptual and physical geographies between genres. The EPFOI website (launching this spring) displays these visualizations for scholars to consider for their own research.
In addition to the visualizations, the website includes what we call “micro essays,” or short blurbs that describe trends we have seen in the visualizations and possible reasons why these trends occurred. These micro essays are tools to facilitate “mid-range” reading, and although they are not fully fleshed out, they offer scholars potential avenues for further research. The website also provides pedagogical tips for those who would like to bring the French of Italy into the classroom. The Center for Medieval Studies has further shared its data for EPFOI on the website and invites scholars to use it for their own research.
This project showcase will explore the EPFOI website and describe the process of developing the project. Additionally, it will consider how EPFOI may serve as a methodological model for “mid-range” reading of other small corpora from any historical period.
PA Digital and the DPLA: Collaborating on Collections as a Community
Patricia Hswe
Presentation Slides
This project showcase will provide an update on the Pennsylvania Digital Collections Project (PDCP) and its formation of a service hub for the Digital Public Library of America (DPLA). Launched in 2013, the DPLA is consolidating digital collections through a common portal, exposing primary sources, ripe for humanities teaching and research, on a scale similar to other national digital library efforts such as HathiTrust. The PDCP was organized in 2014 to provide a more cohesive framework for statewide collaboration on enhancing discoverability of, and access to, cultural heritage resources unique to the Commonwealth. As a community of collaboration, it has mobilized participant institutions around the mutual goal of building support, service, and technology infrastructure for the sharing of Pennsylvania digital collection content - and much more broadly than would otherwise be possible as individual institutions.
The project showcase will focus on the initiative that kicked off PDCP planning and organization early on: its engagement in the DPLA as a service hub for the Commonwealth. The presentation will highlight analysis of responses to a survey on digital collections, deployed to determine statewide readiness for participation in the PDCP for the DPLA; metadata guidelines for digital collections to be harvested; and the development and implementation of the Hydra-based PDCP Aggregator, used for harvesting metadata from the contributing organizations to push out for sharing via the DPLA.
Besides reporting out on progress, the purpose of the showcase is also to present a model of collaboration for a national effort - the DPLA - that is enlarging upon usage, manipulation, and application of collection content for, among other types of users, digital humanities practitioners. In doing so, the session will also address challenges, lessons learned, and anticipated future PDCP endeavors relative to the Project’s continuing involvement as a service hub for the DPLA.
Six Degrees of Francis Bacon: A Project Showcase
Jessica Otis
Six Degrees of Francis Bacon is an interdisciplinary digital humanities project, reconstructing the social network of early modern Britain (1500-1700) through a combination of machine learning and scholarly crowdsourcing. This project showcase will highlight the collaboration of professors and students in Georgetown University’s Department of English and Carnegie Mellon University’s Departments of English and Statistics; Information Systems Program; Hunt Library; and Human-Computing Interaction Institute, to statistically infer the Six Degrees network then display that network on a website suitable for crowdsourced input.
The Six Degrees team employed a combination of freely-available Named-Entity Recognition tools and custom statistical inference on the Oxford Dictionary of National Biography to create a preliminary social network of over 13,000 early modern people and 170,000 relationships between them. This inferred network was not intended to be a definitive statement of who knew whom in early modern Britain, but rather the foundation for expert intervention and crowdsourcing. To facilitate this, the Six Degrees team designed a series of prototype websites–iteratively improving our design–before finally going live with our beta website in September 2015. To date, the site has received approximately 30,000 hits from over 3,000 users, who have collectively made over 5,100 contributions to our preliminary network.
The Restoration of Nell Nelson
Rob Spadafore and Rebecca Parker
Helen Cusach’s muckraking that exposed the unsanitary working conditions and cruel mistreatment of the women and children in Chicago’s manufacturing industry was published under the pseudonym Nell Nelson in her series “City Slave Girls” printed by The Chicago Times in late July 1888. Cusach’s satirical and witty exposé created an immediate social agitation causing some of Chicago’s most prominent men to publicly react. This included responses from Chicago’s governor John Altgeld, the president of the Chicago Board of Trade Charles Hutchinson and the vice president of the First National Bank Lyman Cage, among others. Owners of the textile mills, meat-packing factories and the other commodity industries that Cusach scrutinized recognized the threat, with one manufacturer ultimately suing the paper for misrepresentation. Later in 1888 the Barkley Publishing Company printed an abridged version of the newspaper series in a book titled City Slave Girls. Author John McEnnis also included parts of the series in his 1888 pro-union book The White Slave Girls of Free America. Reprinted in the New York World and referenced by labor and women’s rights groups, such as the Illinois Woman’s Alliance, the series gained nationwide coverage. Yet, presently Nelson’s series is infrequently referenced in the discussion of American industrialization or investigative reporting; her work superseded by that of Upton Sinclair and Nellie Bly.
The Restoration of Nell Nelson project aims to change this through the digitization of Cusach’s series. We are making the entire text of the articles available as searchable TEI constrained XML. Using XSLT, SVG and JavaScript we highlight Cusach’s distinct use of language and her representation of gender and archetypal speakers. By comparing the original series with the republications we plan to indicate how changes in Cusach’s language altered the series original intent for complete exposure of corrupt Chicago industrialization.
The Luther Works Visualization Project
Chuck Steel
The Luther Works Visualization Project is meant to provide a resource for both Luther scholars and lay persons with an interest in Luther’s works to discover and explore them through the use of word clouds. While it should not be considered a tool for providing deep scholarly research on Luther’s works it can provide insights that may point the scholar in a direction they had not considered in their previous work. The visual nature of word clouds can also provide the lay person with a quick synopsis of a group of Luther’s works, or even of a single text, that will help them grasp the key themes of the texts in question. This can make the works more approachable for those that are just beginning their journey with Luther. I developed this project to fulfill the requirements of my Master of Arts in Religion degree at the Lutheran Theological Seminary at Gettysburg. The project database includes over six hundred works by Luther including letters, sermons, hymns and other writings the majority of which are in the public domain. The website provides an interface that allows the user to explore Luther’s works in three different ways; browsing works by type, exploring how Luther’s use of words changed over time, and performing a search using various criteria such as type of work, date, keywords used. Browsing by type and the search pages also allow the user to compare groups of works to each other. The website is open to the public and can be viewed here: http://projects.chucksteel.com/luther/. My project showcase would give an overview of the web technologies used in creating the project and a demonstration of the website’s features.
Mapping the Imagined South: GIS Mapping of Contemporary Southern Cookbooks
Carrie Tippen, Lisa Cuyler, Kaitlyn Shirey, Tessa Webber, and Rachel Geffrey
Presentation Slides
This project showcase is a presentation of an ongoing project mapping place names that occur in 21st century cookbooks from the American South.
It has long been a truism in American and Southern studies that the boundaries of “The South” of popular imagination have never fit comfortably on the geographic borders of the states of the Confederacy. Texts like Jennifer Rae Greeson’s Our South: Geographic Fantasy and the Rise of National Literature suggest that “This South that we hold collectively in our minds is not – could not possibly be – a fixed or real place” (Greeson 1). And yet, contemporary cookbooks of Southern cuisine appear convinced that such a place does exist, that it is distinct from other places as a culture, and that its borders can be known. Cookbooks trade in the currency of “authenticity” which depends on an image of a continuous geographic South and authentic Southern identity.
To explore this phenomenon of an imagined Southern geography, I employ methods of digital humanities research and GIS technology to draw maps of the South as imagined by each cookbook. The maps represent the breadth of global influences in contemporary Southern cuisine by tracking explicit references to the origins of ingredients or Southern food traditions. The maps also represent the centers of Southern cuisine through “density” or “heat” mapping, marking repeated references.
Though the project is ongoing, this presentation focuses on two maps that demonstrate dramatically different Souths: John Currence’s Pickles, Pigs, and Whiskey (2013) and Sean Brock’s Heritage (2014). The maps show that Currence’s experiences living and cooking in New Orleans and Mississippi yield a very different map from Brock, whose experience is centered in rural Virginia, South Carolina, and Tennessee. Though both cookbooks appear to represent the authentic cuisine of an Imagined South, the maps reveal that they represent relatively small – and scarcely overlapping – portions of the geographic South.
Taking Livingstone Online across Disciplines, Institutions, and Continents
Adrian Wisnicki and Ashanka Kumari
This project showcase will explore the role of international, interdisciplinary collaboration in two linked, NEH-funded projects: the Livingstone Spectral Imaging Project (2010-present) and the Livingstone Online Enrichment and Access Project (2013-present). Collectively, these projects bring together over 20 senior and junior specialists, inside and outside the academy, spread across three continents (North America, Europe, Africa) as well as nearly 40 supporting institutions from around the world. Expertise ranges across areas such as literature, history, writing studies, medicine, physics, computer science, library science, data management, and web design.
Both projects focus on David Livingstone (1813-73), the celebrated Victorian traveler, missionary, and abolitionist. The spectral imaging project applies advanced imaging technology to study the textual and material history of some of Livingstone’s most damaged manuscripts. Livingstone Online centers on building a visual and textual archive of Livingstone manuscripts and artifacts, with an emphasis on restoring Livingstone – an otherwise iconic and oft dehistoricized figure – to the non-Western cultural contexts in which he worked and traveled. Each project places special emphasis on accountability and knowledge transfer via extensive documentation in the form of narrative text, images, and downloadable documents. Such documentation also makes transparent the inner workings of our collaborative relationships.
This showcase will walk through the role of collaboration in enabling our work while touching on facets such as project planning, program management, division of labor, communication among team members and across disciplines, and reporting to stakeholders. We will argue that the spatial distribution of our collaborators and the division of our team into overlapping sub-units (scholarship, site design, data development, outreach and dissemination, management) enables us to work at an exceptional scale and, advantageously, to integrate diverse perspectives into our research. However, we will also cite the challenges that such scale and distribution poses and explore the potential conflicts that arise when scholar-led, people-oriented projects such as ours must negotiate their way within broader institutional – particularly university – contexts.