University of Pittsburgh

Overview

This first NA+DAH event was designed to introduce the teams to one another and to get the participants up to speed on working with their data, thinking in networks, and managing their projects. The week-long workshop was designed with a scaffolded approach, moving from establishing a shared foundation, to learning how to best prepare data, and then focusing on analyzing that data through network analystic techniques. Each day had workshop time, group work time/office hours to allow participants to gain guidance directly from the invited experts, keynote speakers, and technical assistants. Each of these components were integrated together so that, for example, the keynotes were valuable contributions to the daily conversations, and the workshops advanced the teams’ projects. Critical discussion was invited throughout. 

The keynote presentations, delivered by leading experts (see below), were intentionally distributed over the course of the week to support this learning arc and scheduled to ensure that the community could have adequate time to digest and reflect upon the content in dialogue with the speaker. The keynotes also offered a balance between theoretical and applied approaches.

The workshops were 90-minute, hands-on sessions (see below). Topics included: intermediate-to-advanced approaches to networks, nodes, and edges; how to model data in flat-file, relational and graph database structures; tools for visualizing and manipulating networks and networked data (OpenRefine, Gephi, etc…); advanced network concepts (bimodal networks, etc.); and how to create and sustain interdisciplinary teams.

We found it effective to focus this multi-event advanced digital humanities workshop on teams rather than individuals. As evidenced by team conversations and work that stretched into every available hour, even times dedicated to social events, the teams savored and valued the time to work together. After a week of working with our keynote speakers, workshop leaders, and one-on-one with our technical assistants, participants were able to present successfully a more detailed project scope, a strong understanding of how they need to revise their data, and a one-year plan for preliminary data manipulation, analysis, and further development of their research inquiry.

To view the original schedule, feel free to click here.

Keynote Synopses

• Ruth Ahnert, “Early Modern Letter Networks”
  
  • This talk introduced the analysis of three different early modern letter collections, and discussed the benefits and challenges of working with data at different scales and with different kinds of provenance. It touched on issues of data cleaning, quantitative analysis, and reconciliation of metadata from different archives. 
• Pamela Fletcher, “Digital Dividends”
  • Are there compelling art-historical reasons to adopt digital and computational methods? The allure of going digital for some scholars and institutions is matched by the skepticism evinced by others, who see digital work as, at best, peripheral to art-historical questions and at worst, actively hostile to them. In this talk, Fletcher used her own experiences with a long-term digital project and with the development of a program in digital and computational studies as a starting point to reflect on how art-historical research agendas intersect with computational work, and to prompt discussion about how individual projects locate themselves within the larger questions and discourses of the discipline.
• Charles van den Heuvel, “Small Worlds and the Big Picture: Creating Networks for Art Histories”
  • In the project: Golden Agents. Creative Industries and the Making of the Dutch Golden Age, a consortium of Dutch academic and cultural heritage institutions is developing a research infrastructure combining semantic web and multi-agent technologies to analyze interactions between the various branches and the production and consumption of (im-) material cultural goods in the Low Countries in the 17th and 18th Centuries, in particular in Amsterdam. 2,000,000 scans of probate inventories of households in Amsterdam are disclosed to get a better understanding of the consumption of cultural goods of the Dutch Golden Age. The aim is to create a big picture of the Dutch Golden Age and to go beyond research questions that are often based on extrapolations of studies of the elite-culture of Amsterdam in the 17^th century. To make this big picture, all sorts of networks need to be (re-)created not only for visualization but also for data curation. One type of network concerns the creation of small worlds of networks of people to disambiguate and to identify persons from the big data of names in the City Archives of Amsterdam and to link these to metadata of other academic and cultural heritage institutions. On a more generic level, van den Heuvel also discussed working in a distributed network of (inter-)national institutions with different policies and missions as well as the impact of the use of big data for skills and expertise needed in digital research methods of art histories.
• David Newbury, “Summary”
  • This presentation, offered at the very end of the week, synthesized the work and conversations that took place throughout the convening, drawing connections between the various projects, workshops, and keynotes from the perspective of a technologist and cultural heritage professional.
• Cosma Shalizi, “Network Analysis: A Network Analysts Perspective”
  • This keynote introduced the assembled teams to the state of the field of network analysis itself, along with the leading questions/challenges in that field. Shalizi presented a view into digital art history from the network analytic point-of-view.

Workshop Synopses

• Network Basics (Ruth Ahnert)
  • This workshop introduced networks, nodes, edges, directed and weighted networks, bi- and multi-partite networks. It gave an overview of the kinds of things that can be thought about through a network framework, as well as some things that can’t, and it will introduce key theories, including weak ties, and small worlds. Ahnert presented an activity where participants built their own test data set that formed the basis of the activities in Nicole Coleman’s first workshop, “Transforming and Visualizing Data,” as well as Ahnert’s second workshop, “Getting Quantitative.”
• Getting Quantitative (Ruth Ahnert)
  • This workshop covered network metrics including various centrality measures, clustering coefficients, and community detection algorithms. It included an activity that allowed the participants to run through some of these algorithms and provided suggestions for routes forward with other tools and coding libraries that allow quantitative analysis.
• Transforming and Visualizing Data (Nicole Coleman)
  • This workshop will use the test data set created in the “Network Basics” workshop as a starting point for manipulating and extending project data as research progresses. The workshop introduced tools and methods for collaborative data editing as well as managing data over the course of a project. Activities included the use of OpenRefine to explore tabular data and also transform data for use in visualization. The introduction to OpenRefine will lay the groundwork for more advanced use in Coleman’s second workshop, “Linked Data and Reconciling Authorities.”
• Visual Analyses of Network Data (Nicole Coleman)
  • This workshop introduced techniques for visualization that go beyond nodes and edges and provided informationon, and practice with, what can be learned by exploring data in a range of different views.
• Network+Map (Weixuan Li)
  • The application of network analysis in art historical research often equates “network” to “social network” and borrows concepts and theories from the social sciences. Digital art history, however, calls for new concepts tailored to artistic networks in order to account for the developments of art in all its complexity. This workshop will propel participants to think and conceptualizes a novel art historical network of ideas, such as that of iconography, connecting artists not through social ties but shared subject matters. During this interactive workshop, participants received hands-on experience constructing and visualizing the conceptual, art historical network according to their interests. When the network has geographical features, the participants also learned how to demonstrate the network together with the maps of their choice.
• Linked Art Historical Data (Matthew Lincoln)
  • Almost all art history research projects involve objects and documents held by external institutions, such as archives, libraries, and museums. We have the opportunity to deeply link our novel, scholarly research data with the collections data produced by these cultural heritage institutions, but huge “Linked Open Data” projects often seem to require an army of highly-paid developers that we can’t afford. In this workshop, Lincoln talked about a middle path for real-life researchers trying to get the most out of connecting their newly-created knowledge (and the data that come with it) back into the larger ecosystem on which we all depend.
• Missing Data and Uncertainty (Matthew Lincoln)
  • Uncertain historical data is like the weather: everyone talks about it, but no one does anything about it. In this session. Lincoln discussed common missing/uncertain data issues in historical datasets, special considerations for missing data when it comes to network analysis, and strategies for encoding, acknowledging, and even measuring uncertainty in your data.

Presenters

Ruth Ahnert, Senior Lecturer [Associate Professor] in Renaissance Studies, Queen Mary University of London; Fellow of the Alan Turing Institute; Principal Investigator on the project Living with Machines; and Co-Investigator on Networking Archives.

Nicole Coleman, Digital Research Architect, Research Director, Humanities+Design, Stanford Libraries, Stanford University

Pamela Fletcher, Professor of Art History, Bowdoin College. Fletcher is the author of Narrating Modernity: The British Problem Picture 1895-1914 (2003); the co-editor (with Anne Helmreich) of The Rise of the Modern Art Market in London 1850-1939 (2011); co-creator (with David Israel) of The London Gallery Project (2007, revised 2012); and is currently writing a history of Victorian modern-life painting. She was a founding co-director of the Digital and Computational Studies Initiative at Bowdoin; served as the first editor for digital humanities and art history at caa.reviews; and is currently Vice President for Programming, Digital Art History Society.

Charles van den Heuvel, Head of Department of History of Knowledge, Huygens ING (KNAW); Professor of Digital Methods and History at the University of Amsterdam.  van den Heuvel has a background in history of art, specializing in the history of town planning, fortification and architecture of the Early Modern Period and has worked in several cultural heritage institutions. Recent interests include the digital humanities (in particular spatial humanities), the history of knowledge (in particular the Digital Republic of Letters), and the history of library and information sciences (in particular the history of classification).

Weixuan Li, PhD candidate in the project “Virtual Interiors as Interfaces for Big Historical Data Research” at Huygens ING. Her research focuses on comprehending the relationship between the urban fabric and artists’ location choice within Golden Age Amsterdam. Combining her training in digital methods at MIT and art history at the University of Amsterdam, she worked on visualizing the art and art market in the early modern Netherlands from different perspectives. Her previous projects include ‘Visualizing the spread of Ideas,’ and ‘Constructing a network of artistic ideas through Iconography.’

Matthew Lincoln, Digital Humanities Developer, at Carnegie Mellon University Libraries, where he focuses on computational and data-driven approaches to the study of history and culture. He earned his PhD in Art History at the University of Maryland, College Park, and has held positions at the Getty Research Institute and the National Gallery of Art. He is an editorial board member of The Programming Historian.

David Newbury, Enterprise Software Architect at the J. Paul Getty Trust, where he works with museum professionals, researchers, scientists and technologists on to find common solutions to technical and scholarly problems. He has also worked on other collaborative museum technology projects such as Art Tracks, a provenance project at the Carnegie Museum of Art, and the American Art Collaborative, working with 14 museums on standardising models and software around Linked Data. He has previously worked with Carnegie Mellon University, the University of British Columbia, University of Illinois, and PBS.

Cosma Shalizi, Associate Professor, Statistics Department, Dietrich College of Humanities and Social Sciences, and Associate Professor, Machine Learning Department, School of Computer Science, Carnegie Mellon University; External Professor, Sante Fe Institute

Technical Assistants

S. E. Hackney (they/them/theirs), PhD candidate in Library and Information Science at the University of Pittsburgh’s School of Computing and Information, MSLIS with Advanced Certificate in Digital Humanities from Pratt Institute’s School of Information. Their research focuses on structural inequality within digital infrastructure systems, particularly within the realm of digital character-encoding standards, and the ways that knowledge organization systems create physical and virtual spaces that privilege certain bodies and experiences over others.

John Ladd, Postdoctoral Fellow, Kaplan Institute for the Humanities, Northwestern University (beginning Fall 2019); former research fellow for Six Degrees of Francis Bacon, Carnegie Mellon University; PhD in English Literature from Washington University in St. Louis.

Project Associates

Sarah Reiff Conell, PhD candidate in the History of Art and Architecture Department at the University of Pittsburgh, employs digital methods, including network analysis, to investigate late medieval and early modern cults of the Virgin Mary in the Danube River Valley. Her research traces the flow of miraculous agency through objects that participate in various forms of replication both within and across media.

Meredith North, PhD candidate in the History of Art and Architecture Department at the University of Pittsburgh, specializes in Modern and Contemporary German artistic practices. Other research interests include transnational artistic, social, and political connections between the United States and Germany in the Cold War era, queer identity and [mis]representation, and the history of intermedia art and immersive environments.