Digital Methods

Learning Outcomes

Much of the subjectmatter in this course is driven by student interests, unbound by a shared commitment to any dingle topic or even discipline. What follows serves as a shared reference point for the feedback you receive on assignments and exercises, and as a basis for your final report, where you assess your own work (see (Un)Grading).

Research and Modeling

  1. Ask a research question about a text, or a body of texts, that computational methods can help answer, and explain why it matters in your field.
  2. Analyze a document: identify its structures and features, and decide which of them to model, justifying your choices in terms of your research question.
  3. Design a schema that expresses your model, and explain what it allows and what it rules out.

Coding Literacy

  1. Explain core computational concepts (paths, loops, conditionals, data types, and functions) in plain language, and demonstrate them on paper.
  2. Read code you did not write, including XML (TEI or otherwise), regular expressions, XPath, XSLT, and Python, and explain what each part does and why it is there.
  3. Mark up a text, both by hand and with AI-assistance, and check it against your schema.
  4. Extract information from a structured text and turn it into something new: a reading edition, a table, a chart, a network, or a map.

Working with AI

  1. Explain in general terms how large language models work, including the difference between deterministic and stochastic systems, and what follows for when to trust them.
  2. Specify a task precisely enough for an AI tool to carry it out, and recognize when it has not; learn to build effective stylesheets and spec documents.
  3. Take responsibility for the code you use: verify it independently, by reading it, by testing it on cases with known answers, and by trying to break it, never by taking the model’s word for it.
  4. Look things up: find answers you do not already have, in reference documentation, online, and from AI, and judge which of those answers to trust.

Research in Practice

  1. Present your research in a form other people can see and use, including website design.
  2. Collaborate: divide the work, communicate, credit each other’s contributions, learn workflow and version control best practices.
  3. Reflect on your own learning in writing, and in a way that feeds back into the research process.
  4. Discuss how digital methods, and AI in particular, are changing research in the humanities, drawing on other digital humanities projects.