Recommended
More Related Content
What's hot
What's hot (20)
Viewers also liked
Viewers also liked (12)
Similar to Humanities as Data: Projects, Visualizations, and Emerging Methods
Similar to Humanities as Data: Projects, Visualizations, and Emerging Methods (20)
Recently uploaded
Recently uploaded (20)
Humanities as Data: Projects, Visualizations, and Emerging Methods
- 1. Humanities as Data Projects, Visualizations, and Emerging Methods Kurt Fendt Massachusetts Institute of Technology fendt@mit.edu @fendt hyperstudio.mit.edu @MIThyperstudio
- 2. Outline • Digital Humanities • Short History • Trends • New Affordances • HyperStudio - Digital Humanities at MIT • Structure, Principles • Selected Projects • Berliner sehen • Annotation Studio & Open Source • Data Visualization - The Comédie-Française Registers Project • Parallel Axis Graph • Combinatorial and Generative Research Visualization Tools • Network Graphs • Educating Digital Humanists • Project-Based Digital Humanities Course at MIT • Q & A hyperstudio.mit.edu @MIThyperstudio
- 3. Digital Humanities - A Definition • The digital humanities, also known as humanities computing, is a field of research, teaching, and invention concerned with the intersection of computing and the disciplines of the humanities. • It is methodological by nature and interdisciplinary in scope. • It involves investigation, analysis, synthesis and presentation of information in electronic form. • It studies how these media affect the disciplines in which they are used, and what these disciplines have to contribute to our knowledge of computing. Wikipedia, s.v. „Digital Humanities“, last modified July 31, 2011, http://en.wikipedia.org/wiki/Digital_humanities hyperstudio.mit.edu @MIThyperstudio
- 4. Digital Humanities - A Brief History hyperstudio.mit.edu @MIThyperstudio Father Busa
- 5. Digital Humanities - A Brief History hyperstudio.mit.edu @MIThyperstudio Vannevar Bush: “As We May Think” (1945)
- 6. Digital Humanities - A Brief History hyperstudio.mit.edu @MIThyperstudio Theodor H. Nelson: “Literary Machines” (1965/1981)
- 7. Digital Humanities - “The Big Tent” hyperstudio.mit.edu @MIThyperstudio
- 8. Digital Humanities - Trends hyperstudio.mit.edu @MIThyperstudio Big Data: Mapping the Enlightenment The Electronic Enlightenment database contains over 55,000 letters and documents exchanged between 6,400 correspondents in the Republic of Letters. How can humanities scholars trained in close reading of individual documents make sense of patterns in large sets of data? How can historians and other humanities scholars use visualization tools, to examine large sets of heterogeneous historical data with multiple dimensions? http://www.stanford.edu/group/toolingup/rplviz/
- 9. Digital Humanities - Trends hyperstudio.mit.edu @MIThyperstudio Data Visualization http://www.visual-literacy.org/periodic_table/periodic_table.html#
- 10. Digital Humanities - Trends hyperstudio.mit.edu @MIThyperstudio Annotation
- 11. Digital Humanities - New Affordances • Asking and answering new research questions that cannot be reduced to a single genre, medium, discipline, or institution • New research methods, representational and interpretive practices, meaning-making strategies, complexities, and ambiguities • Fluid communities of practice • Trans-historical and transmedia approach to knowledge and meaning- making • Questions of design at the center (information design, graphics, typography, formal and rhetorical patterning) • Project as the core activity “A project is a kind of scholarship that requires design, management, negotiation, and collaboration.” Anne Burdick et al.:„Digital_Humanities“, Cambridge, MA 2012, MIT Press hyperstudio.mit.edu @MIThyperstudio
- 12. HyperStudio - Digital Humanities at MIT hyperstudio.mit.edu @MIThyperstudio
- 13. Berliner sehen - Database Narrative for Education hyperstudio.mit.edu @MIThyperstudio
- 15. HyperStudio as part of Comparative Media Studies hyperstudio.mit.edu @MIThyperstudio • One of nine independent research groups within the Department of Comparative Media Studies/Writing (CMS/W) (School of Humanities, Arts, and Social Sciences) Other CMS research groups include: Center for Civic Media; Education Arcade; E-Lab; Imagination, Computation, and Expression Lab; MIT Game Lab; Open Documentary Lab; Mobile Experience Lab; Trope Tank • Concept of Applied Humanities (Henry Jenkins) • MIT Motto: Mens et Manus • HyperStudio: 9 part-time and full-time staff (Graduate/undergraduate students, software engineers, outside contractors, administrator)
- 16. HyperStudio - Principles hyperstudio.mit.edu @MIThyperstudio • Pedagogical and/or scholarly needs drive development • Co-design with faculty, students, and other partners • Agile development with integrated feedback • Students as novice scholars • Engage learners in process of discovery, interpretation, and collaboration • Rethinking of pedagogical concepts and roles
- 17. Multimedia Text Annotation for Students “I have never annotated before. But I think I am getting better. I am actually writing down ideas while reading. By writing them down, I am actually looking deeper into the text, not like when I just read the book or something and said, ‘Oh it may mean this.’ Now it is more like, ‘Oh what does THIS mean?’ Then I keep asking questions because I am annotating. I am thinking about the text more.” Student in a Fall 2012 literature class hyperstudio.mit.edu @MIThyperstudio
- 18. Iliad: Venetus, 10th century hyperstudio.mit.edu @MIThyperstudio
- 21. Pedagogical Approach • Increase awareness of fluid processes of reading, writing, borrowing, and revision (John Bryant) Engage students as “editors” (Wyn Kelley) Develop traditional humanistic skills (close reading, textual analysis, persuasive writing, critical thinking) Allow students to practice “scholarly primitives” “ I’m using the term “primitives” in a self-consciously analogical way, to refer to some basic functions common to scholarly activity across disciplines, over time, and independent of theoretical orientation.“ John Unsworth Discovering Annotating Comparing Referring Sampling Illustrating hyperstudio.mit.edu @MIThyperstudio
- 22. Annotation Studio Multimedia Text Annotations for Students hyperstudio.mit.edu @MIThyperstudio
- 23. Annotation Studio Multimedia Text Annotations for Students hyperstudio.mit.edu @MIThyperstudio
- 24. Annotation Studio Multimedia Text Annotations for Students hyperstudio.mit.edu @MIThyperstudio
- 25. Annotation Studio Multimedia Text Annotations for Students Annotation • Citation (reference to base text plus metadata) • Comment • Tags (folksonomies) • Links to other sources • User information (name, group) • Date/time hyperstudio.mit.edu @MIThyperstudio
- 26. Annotation Across Classroom Practices • Close Reading • Generating Material: 800 comments on one text • Developing an Argument • Revision • Research and Presentation • Making Connections Across Texts • Peer Review and Social Reading • Reflecting on Processes of Reading, Writing, and Sharing Work hyperstudio.mit.edu @MIThyperstudio
- 27. Future Directions • Incorporating student-generated texts for annotationDeveloping citation toolsFiltering annotations by student, subject, date, etc. • Exporting annotations into visual drafting spaceSupporting creative writing and translation courses • Side-by-side display of texts/media documents • Annotation across multiple documents • Annotation of multimedia sources (image, video, audio) • Customizable visual display of annotations • Curated repository of media and text documents • Export and archiving of annotations (Open Annotation Standard) • Connection to other tools via open API • Version for mobile devices hyperstudio.mit.edu @MIThyperstudio
- 29. Annotation Studio Multimedia Text Annotations for Students hyperstudio.mit.edu @MIThyperstudio
- 30. Open Source: New Opportunities hyperstudio.mit.edu @MIThyperstudio • Annotation Studio based on the Annotator by the Open Knowledge Foundation • Annotation Studio code open source as well (GPL 2) • Rich community of developers • Other groups can fork code, contribute, build upon • Use of open APIs (application programming interfaces) allows new forms of collaboration, e.g. visualization tools, filter mechanisms • Annotation Studio can be freely installed or run as a service • Basis for other projects, e.g. Lacuna Stories at Stanford U., Hofstra University, and New York University • Used by almost 150 institutions of higher education in the fall • Open source is a requirement by the National Endowment for the Humanities (federal US funding agency)
- 31. Comédie-Française Registers Project hyperstudio.mit.edu @MIThyperstudio Project Director: Prof. Jeffrey Ravel, MIT History Department
- 32. The Comédie-Française Registers Project The Comédie-Française Registers Project (CFRP) is a web resource for scholars of 17th & 18th c. French theater to support an exploratory research process. Three Components •Archive •Search tool (faceted browser) •Interactive data visualization tools hyperstudio.mit.edu @MIThyperstudio * Original version of the CFRP slides were created by Jason Lipshin
- 33. Merging Domain Expertise + DH Methods hyperstudio.mit.edu @MIThyperstudio
- 34. The CFRP Online Archive Significance for domain experts (French theater historians): • Digitized access to rare materials • Cultural significance of the time period (i.e. the French Revolution) • Granular search through extensive archives. hyperstudio.mit.edu @MIThyperstudio
- 35. Data Visualization as Methodology For digital humanists: data viz as part of an exploratory research process. • ”Machine Reading” (Ramsay) – Macro-level analysis and the affordances of computation enabling new research questions. • “Toggling” (Schnapp et all) – Merging quantitative and qualitative analysis of historical data. • Combinatorial Research – Dynamically combining parameters as generative analysis. hyperstudio.mit.edu @MIThyperstudio
- 36. Archive and Document Dimensions The Comédie-Française Archive: 1680 – 1793 • Daily records of repertory and box office receipts • Information on actors’ roles, payment, and playwrights • Daily Expenses Register Elements: • Play title • Author • Actors • Year • Number of tickets sold • Ticket price • Location of seats in theater • Premiere, first run, or revival hyperstudio.mit.edu @MIThyperstudio
- 37. Archive and Document Dimensions The Comédie-Française Archive: 1680 – 1793 • 113 Seasons • Approximately 320 records per season • 2 plays per day • 4 genre categories Data challenges and difficulties: • Troupe occupies 4 different theaters • Each theater has between 5 and 7 sections • These sections translate to between 13 and 21 ticket price categories hyperstudio.mit.edu @MIThyperstudio
- 38. Visualization: Case Studies Parallel Axis Graph • Dynamic relations between all categories recorded in the registers Theater Mapping • Diagram of theater layout acts as navigation to the database. Line Graph (Voltaire’s Mahomet) • Tracing the history of one play throughout its performance and reperformance. Network Graphs • Repertoire decisions, popularity of plays hyperstudio.mit.edu @MIThyperstudio
- 39. Visualization: Case Study 1 Parallel Axis Graph • Dynamic relations between all categories recorded in the registers and external events hyperstudio.mit.edu @MIThyperstudio
- 40. Visualization: Case Study 2 Theater Mapping • Diagram of physical space of theater acts as navigation to the database. hyperstudio.mit.edu @MIThyperstudio
- 41. Visualization: Case Study 3 hyperstudio.mit.edu @MIThyperstudio
- 42. Visualization: Case Study 3 Mahomet ou le Fanatisme: Voltaire •Tracing the history of one play throughout its performance •72 instances of Mahomet within the 13 year period between 1780 and 1793. hyperstudio.mit.edu @MIThyperstudio
- 43. Future Directions The Iterative Design Process •New visualizations based on research questions from domain experts. •Collaborations with new scholars and institutions: Comédie-Italienne, Broadway, Opéra de Paris. •Generalizability of tools to other kinds of data. •New browser tool for dynamic visualization creation (Chris Dessonville) •Network Graphs to explore repertoire decisions hyperstudio.mit.edu @MIThyperstudio
- 44. New Browser Tool Combinatorial and Generative Research •The user can select which parameters to compare and the system will automatically generate a list of potential visualizations. •The visualization will load in the same facet without the need for refreshing. hyperstudio.mit.edu @MIThyperstudio
- 48. Project-Based Digital Humanities Course hyperstudio.mit.edu @MIThyperstudio • Thirteen 4-hour course units, 17 Graduate and Undergraduate Students from Computer Science, Art History, Architecture, CMS, Mechanical Engineering (MIT, Harvard University, Wellesley University, Mass. College of Art) • Each unit included: • discussion of readings and introduction to new topics /guest speakers • small data/tool experiments • discussion and work on larger group projects • Four larger group projects primarily with outside partners: Institute for Contemporary Art, Isabella Stuart Gardner Museum, MIT Museum, Comédie-Française Registers Project
Editor's Notes
- 2004 ALLC/ACH Conference in Gothenburg, Sweden – Conference renamed to Digital Humanities Conference Indovidual scholar, used computers for the first time, concordance, literature as data
- other beginnings that are equally imprortnat if not more important to what we describe now as digital humanities, Dean of Engineering at MIT, used the mind as a model to rethink the way we process information, namely by association, linking, juxtaposition, Google Glass
- coined the term Hypertext, literature as the model (Talmud), palimpsest
- now big tent, recent publications
- one out of many examples, often connected to visualization, see also Lev Manovich’s work on Culturomics, taking large datasets such as Manga comic book covers and visualizing them.
- Meta slide, visualization of the visualization methods, look at the URL: Visual Literarcy
- Annotaion, DH 2012 in Hamburg
- Focus on scholarship, research, So how does this all play out? Effect on learning and the training of novice scholars has not been the focus (more on this a bit later when I talk about the DH course) That's where HyperStudio comes in
- Talk about two projects (out of about 35 that we have done so far)
- So if I were to describe the project in elevator pitch form, I would say that CFRP is a web resource for researchers of 18th century French theater positioned at the intersection of quantitative and qualitative approaches. This resource is really a constellation of three very integrated components: It consists of a very large online archive of hundreds of years of documents related to France’s national theater troupe, the Comedie Francaise. A faceted browser – for granular search of individual documents within that archive. And perhaps most excitingly, a series of interactive visualization tools, which we are currently in the process of prototyping and which I’ll discuss in a little more detail in a little bit.
- But before I get into the nitty-gritty and specific details about what the project entails
- Socioeconomic makeup of the audience. Which plays were popular with certain classes? - Tickets in the loges were much more expensive than tickets in the parterre.
- Faceted browser and search is tightly integrated with visualization framework. Can see individual register as well; again, this notion, of toggling from the macro to the micro-point of view. So, for instance, if you were to pick to visualize Voltaire’s YEAR play Mahomet, as it was played from . You can really start to get at this macro-scale perspective.
- Link: http://vimeo.com/53298366 Peak in popularity in 1781-1782 season. What does this peak mean? How does it relate to the themes portrayed in the play – do they resonate with the audience in a charged, new political-economic climate? - Days of the week: how performances at the Comedie-Francaise interacted with other entertainment forms throughout the week. For example, opera was on Tuesdays. How did that effect attendance at the Comedie-Francaise?
- - Whereas many of our previous visualizations were sort of one-off experiments – visualizations tailored to answer specific research questions – this is a tool which allows for a greater deal of free play. The researcher can combine whatever parameters he or she chooses from those available on the registers and the system will dynamically generate a set of potential visualizations (in the same browser window, without having to refresh). This tool comes closest to facilitating what we mean by a truly exploratory research process.
- Using a subset of the data from the CFRP Registers (1769 - 1793), this prototype visualization investigates the relationship between authors and plays which are staged on the same day. Each play or author is represented by a node, the relationship of sharing the same day of performance is a undirected link. Together, this set of network graphs represents approximately 700 unique play titles and 233 authors which cover the 15265 performances put on by the Comedie Francaise over the course of 14 years. The interactive network graphs have features that guide the user from overview to detailed close readings. The user is encouraged to zoom, search, filter, and click through to more information about specific records.These interface features, along with the spatial representation of records work in conjunction to present a series of visualizations guided by 7 research questions:1. Is there a pattern to plays being staged on the same day?2. Are there groups of plays that are more often to be staged with each other?3. Are plays performed the most often also the most profitable?4. Is there a pattern to authors whose plays are staged on the same day?5. Are their groups of authors that are more often staged with each other?6. Are the most popular authors also the most profitable?7. How are central authors such as Moliere’s plays staged in relation to other authors?
- - Whereas many of our previous visualizations were sort of one-off experiments – visualizations tailored to answer specific research questions – this is a tool which allows for a greater deal of free play. The researcher can combine whatever parameters he or she chooses from those available on the registers and the system will dynamically generate a set of potential visualizations (in the same browser window, without having to refresh). This tool comes closest to facilitating what we mean by a truly exploratory research process.
- - Whereas many of our previous visualizations were sort of one-off experiments – visualizations tailored to answer specific research questions – this is a tool which allows for a greater deal of free play. The researcher can combine whatever parameters he or she chooses from those available on the registers and the system will dynamically generate a set of potential visualizations (in the same browser window, without having to refresh). This tool comes closest to facilitating what we mean by a truly exploratory research process.
- - Whereas many of our previous visualizations were sort of one-off experiments – visualizations tailored to answer specific research questions – this is a tool which allows for a greater deal of free play. The researcher can combine whatever parameters he or she chooses from those available on the registers and the system will dynamically generate a set of potential visualizations (in the same browser window, without having to refresh). This tool comes closest to facilitating what we mean by a truly exploratory research process.