1/20/03: Introduction

In this introductory meeting, we will get acquainted, and we will discuss the origin and recent history of the term "knowledge representation," as well as the way it will be applied in this course, in the context of humanities computing. Course requirements will be discussed in detail, in preparation for the first project assignment, due next Monday, January 27th.

Readings:

Drafts from Electronic Textual Editing:

• Mary Case and David Green, Copyright and Permissions in an Electronic Edition

Web:

• Sowa, Knowledge Representation, "Preface"
• Unsworth, Knowledge Representation
• Unsworth, What is Humanities Computing and What is Not?
• Daniel Crevier,"The Tree of Knowledge"

This class is devoted to an overview of the history of computers and computation. Readings (which should be completed before the class meets, for this and all subsequent class meetings) will give long views of the subject (starting with pre-history; starting in the 19th century; starting in the 1950s), and we will discuss our own histories with respect to computers and computation. Project proposals (see below) will also be discussed.

Readings:

Web:

• Howard Rheingold, Tools for Thought, chapter 7, "Machines to Think With"
• UVa Computer Museum
• The History of Computers
• Jon Agar, Lectures on computing: history and culture
• "Probing IBM's Nazi Connection" (CNET interview with Edwin Black, author of "IBM and the Holocaust: The Strategic Alliance Between Nazi Germany and America's Most Powerful Corporation"

Project: Proposals due for a semester-long project. The project needs to include text, images, and maps, and these materials need to be intellectually coherent, as a set, and free of copyright or permissions issues. Complexity is more important than extent, in the materials selected, and students are encouraged to choose something from a period or topic area in which they are interested and feel well versed. (3 pp)

In this section, we will define basic concepts such as "formal" and "algorithm" and discuss the fundamental nature of "computation." Looking at the history of computing, we will talk about the computer as both a conceptual and a mechanical device, focusing on the points at which the conceptual and the mechanical converge and diverge. We will discuss the nature of computation: What is computable? Can a computer generate meaningful inferences?

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Andrea Laue, How Computers Work

Online Reserve Reading:

• Ifrah, Georges. "Binary Arithmetic and Non-Decimal Systems"
• Alan Turing, "Computing Machinery and Intelligence"

Project: On paper, or in a drawing program, model the computation you envision taking place in your project: how will your content be processed by the computer, and what will each part of the computer have to do with that processing. On the software side, consider all aspects of storage and retrieval, transformation, user input, program outputs, etc.. It is not important that you know exactly what parts, or programs, are called--only that you do your best to demonstrate an abstract understanding of how your data and the computer's hardware and software will interact. (1 drawing, 1 page of description)

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Will Thomas, History
  
• Thomas Rommel, Literary Studies
  
• Michael Greenhalgh, Art History

Project: Research the history of humanities computing in the particular area occupied by your project. In a brief written summary, review that history, suggest what can be learned from it in general terms, and specify what can be learned from it in the particular case of your project. (3 pp)

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Michael Sperberg-McQueen, "Classification"

Web:

• Geoff Bowker and Susan Star, "Sorting Things Out: Classification and Its Consequences" (Introduction and Chapter One)

Project: What classification schemes exist, either without or within the world of computing, for the kind of material you are dealing with in your project? Which ones do you think might be relevant? Which of those will you choose to work with, and why? Why not the others? Are there any classificatory schemes that might be adapted from other areas? (3 pp)

As most artifacts studied by humanists are textual, a basic introduction to theories of textuality seems essential. We will trace the history of theories of textuality and practices of editing, laying a theoretical foundation for the unit on digitization and sampling. We will note the historical shift of emphasis from "work" to "text" and will consider current predictions that digital technologies might precipitate the reverse--a return to emphasis on the "work." Attention will also be paid to the structuralist distinction between form and expression with the hope of reinserting rhetoric as a central issue.

Readings: Drafts from Electronic Textual Editing:

• Jerome McGann and Dino Buzzetti, "Editing in a Digital Horizon"

Online Reserve Reading:

• Dino Buzzetti, "Digital Representation and the Text Model"

Web:

• Espen Aarseth, Ergodic Literature
• Allen Renear, David Durand, and Elli Mylonas. "Refining our notion of what text really is: The problem of overlapping hierarchies".

Project: What are the salient characteristics of the textual component of your project? How will you use text descriptions for other parts of the project? Is there an important artifactual layer or carrier medium in your subject matter? How will the characteristics of that artifactual layer be conveyed, if indeed they need to be? (3 pp)

In this unit we will learn the practical skills necessary to digitize artifacts and the critical skills necessary to evaluate the process and products of digitization. We will introduce hardware and software used to digitize artifacts and the data standards used by various text, image and audio file formats. This introduction will focus on the choices involved in digitization and sampling. What is gained and lost in the process? Is it possible to completely represent a material object in digital form? The exercises and subsequent reflection will prompt discussions of best practices in the field of text and image digitization.

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Marilyn Deegan, Conversion of Primary Resources

Drafts from Electronic Textual Editing:

• Dug Duggan and Eileen Fenton, "Effective Methods of Producing Machine-Readable Text from Manuscript and Print Sources"

Online Reserve Reading:

• WJT Mitchell, "Electronic Tools" part 1 and part 2
• Mark Poster, Analogue and Digital

Project: Digitize a sample collection of objects you will use in your project, making sure that you pick both representative and worst-case objects from each data-type, and experiment with alternate methods of digitization where available (OCR vs. transcription of text, for example). We will discuss the tradeoffs between file size, file format, and resolution/color depth, and consider the relationship between digital masters and distribution versions. (at least 15 samples, total)

Most current applications of digital technology rely heavily on structured information, yet the systems of and assumptions underlying structured data are seldom investigated. In this unit, we study these foundations, beginning with the history systems of classification. The intersections of classification systems, philosophy and mathematics will also be investigated. After studying the traditions of classification, we will reflect on our own activities: should humanists transform artifacts into data? If so, how should they structure that data?

Hands-on research activities will prompt additional questions. For instance, what is the appropriate resolution of data? In dealing with dates, how do you record with enough granularity to guarantee precision without producing an information overload that leads to little insight? The benefits of structured data are fully realized only when the appropriate resolution is chosen.

We will also consider the intersections of structured information and research methodologies. Humanities scholars embarking on digital humanities research often worry that in structuring their data they necessarily impose a particular methodology on their research. They worry that structured data will require that they pose problems that are quantifiable. We will consider this conundrum, asking how we might pose problems that are computable without being quantitative.

After considering structured data as an abstract concept we will investigate the two most common systems for structuring data: databases and markup.

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Daniel Pitti, Project Design

Web:

• Daniel Pitti and John Unsworth, "After the Fall: Structured Data at IATH"

Project: Propose different options for structuring your data, and the rationale for each. Most proposals will focus on databases and markup, but other structures might be introduced as well. (3 pp)

• 3/24/03: Databases

  When is a database is the most appropriate tool for structuring information? We will cover the basics of the technology--relational databases, object-oriented databases and SQL--but will also learn to apply a critical eye toward the technologies. Again we will look at the social and intellectual histories of the technology. Who developed these tools and why? What are the intersections of set theory and relational databases? Through the hands-on exercises, we will investigate the complexities involved in actually entering humanities data into a database.

  Readings:

  Drafts from Blackwell Companion to Digital Humanities:

  • Steve Ramsay, Databases
  Web:
  • Frank Stajano, "A Gentle Introduction to Relational and Object-Oriented Databases"

  Project: Sketch the database structure you plan to use in your project, and specify how the database will relate, functionally and conceptually, to the rest of your project. (3 pp)

• 3/31/03: Markup

  One other tool for structuring data, markup technologies are most often used when one sees a hierarchy inherent in the data. There is already much scholarship detailing the advantages and disadvantages of markup. The use of markup in humanities scholarship as also prompted further reflection by bibliographers and textual critics on the natures of text-is it really hierarchical? We will tie this debate to our previous discussions of textuality and ask the question: To what extent have theories of textuality informed theories and applications of markup?

  Readings:

  Drafts from Blackwell Companion to Digital Humanities:

  • Jerome McGann, What is Already Encoded by the Text
  • Allen Renear, Text Encoding and the Fundamental Problems of the Humanities
  Web:
  • Lou Burnard, Michael Sperberg-McQueen, et al., "A Gentle Introduction to XML"

  Project: Sketch the markup structure you plan to use in your project, and specify how the marked up text will relate, functionally and conceptually, to the rest of your project. (3 pp)

The first task in this unit is to examine the cultural authority of maps: we will learn to read maps as interpretations rather than facts. Maps are most often generated to evidence to support an argument. Consistent with other units in this course, we will generate maps that ask questions. We will investigate the coordinate systems that underlay maps. In addition we will investigate perceptions of space in GIS and cartography and, perhaps, Art History. Can we distinguish between space and place?

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Trevor Harris, Maps

Project: Produce at least one map relevant to your project, in at least two ways--as a scanned (raster) image, and as a vector-based GIS map. Visit the Geospatial Information Center in Alderman in search of various types of maps of this place. The goal is to experience maps as structured data plus visualization-or many possible visualizations. We will emphasize the choices and classifications required by each technique as well as the resulting capabilities. (2 maps, 1 page of analysis)

We will learn two basic interpreted programming languages useful for data manipulation: php and xsl. Php is particularly useful for interacting with databases; XSL is useful for manipulating marked up text. Flash may also be an option, for work with maps, timelines, databases, and xml. Through the programming exercises, we will learn some basic principles of programming, such as encapsulation, composition, inheritance, and information hiding. We will also investigate the various social and institutional histories of programming languages.

Readings:

Drafts from Electronic Textual Editing:

• Sebastian Rahtz, Retrieving and Rendering Marked-up Text

Project: Students will learn basic programming through building and populating dynamic web pages using php and/or xsl and/or Flash. (1 working style-sheet, one working php page, or one flash file, demonstrating some previously described functionality)

4/28/03: Interface

In our unit on interface we want to understand the importance of interface within a broader history of computing and to grasp the challenges associated with designing effective and aesthetically appealing interfaces. We also want to broaden the sense of interface, remembering that we are surrounding by interfaces: books, dashboards, ATM machines, stovetops. In the knowledge representation seminar, we will focus on the information architecture side of interface design. How does one organize information in digital environments? How does one balance concerns for audience-ease of use-with concerns for aesthetics-interface as art? How does the metaphor of the "page" affect design in digital environments?

We will also evaluate current efforts to build information architectures and interfaces for projects at institutions such as IATH. In this more focused study we will consider what is involved in making an interface that acts as a provocation. How does one design an interface that will allow users to find what they're looking for AND to discover what they didn't even know existed?

Readings:

Drafts from Blackwell Companion to Digital Humanities:

• Drucker and Nowviskie, Speculative Computing: Aesthetic Provocations in Humanities Computing

Project: Document your interface. Pay particular attention to the balance between audience and aesthetics. For whom was the interface designed? Was it intended to be aesthetically pleasing? Is there evidence of other design constraints? (3 pp.)