Generative AI (GenAI) offers the possibility to support many steps in the creation of digital scholarly editions (DSE), e.g., the production of transcriptions from digital images, the addition of TEI-markup, and the preparation of web presentations (Pollin et al., 2025). There are established procedures and tools for all these steps in the DSE workflow. What is new is that GenAI chatbots make methods easily accessible to users from a single platform via conversation with the bot. With its ubiquity and potential for supporting scholarly work, GenAI should be considered as teaching and learning content in DH (Fritsche and Münster, 2024).

In the summer semester of 2026, a seminar for students of the humanities (German Studies) will be offered at the University of Rostock in which they will work together on a DSE of dramatic works of Sophie Tieck (Dramatische Fantasien, 1804). The seminar runs for 13 weeks, two hours per week. After familiarizing ourselves with the author, her work, and the historical context, we will go through the steps involved in creating an edition and subsequently test how GenAI can help to implement it. A GenAI chatbot using the Gemma 4 model, hosted on a university server, is used, so that there are no constraints for the students. The aim is to create an edition prototype of Tieck’s dramas. Part of the seminar is to discuss and test how GenAI may (1) support the preparation of historical literary texts in digital format, (2) support learning about digital editing methods, (3) be employed in a setting of collaborative work in the classroom using Git. This short paper will present the setup and results of the seminar and reflect on the added value, challenges, and limitations that the use of GenAI brings to DH teaching with regard to digital editions.

How might expanding the range of material formats encoded within a project help draw students’ attention to features of the underlying media in question? Here, collaboration with The Map of Early Modern London (MoEML) holds the potential to shed light. Historically, MoEML’s guidance for encoding primary source transcriptions has been calibrated to parameters of print, as the majority of documents within its project database were initially produced in that medium.[1] Yet the project’s ability to make early modern London’s broader document-based environment accessible and available raises the question of how a range of manuscript texts might best be encoded in relation to the current corpus.

To reflect upon parameters by which students might contribute to work along these lines, as is being envisioned for a Spring 2027 pedagogical partnership with the Massachusetts Institute of Technology, this short paper will focus upon the encoding of indentures as case study, given these manuscript documents’ capacity to serve as both legal and technical infrastructure. Understanding the challenges that this deceptively simple document type may present requires simultaneously considering its early modern significance alongside its potential for facilitating project-specific connections—a dual function that likewise makes the genre crucial for a developing MoEML micro-site, Browsing the Bookshops in St. Paul’s Churchyard, of which students’ encoded documents are planned to become part.

Building upon existing and emerging guidance, including that of the Women Writers’ Project (WWP) and Linked Early Modern Drama Online (LEMDO), the paper will consider both the nature of key challenges currently envisioned and strategies for engaging students as collaborators in the wider endeavor.

[1] For MoEML’s “Encode a Primary Source Transcription” guidance, see https://mapoflondon.uvic.ca/encoding_primary_sources.htm. MoEML also contains a number of born-digital documents, as its project documentation notes.

Text encoding for East Asian and Japanese (EAJ) texts—especially practices guided by the TEI Guidelines—has not diffused widely in Japan, creating a persistent obstacle to sustainable international sharing and cross-corpus research. To address this gap, the SIG East Asian/Japanese, established in 2016, has continued community-based efforts to promote TEI-oriented encoding. Since 2024, this trajectory has been strengthened through a program supported by Japan’s Ministry of Education, Culture, Sports, Science and Technology, bringing international text encoding onto the national research-policy agenda. In parallel, Japan’s National Diet Library has released free OCR software that provides TEI output, indicating that TEI adoption is increasingly supported at an infrastructural level.

At the same time, outside the context of the humanities, Japan has also seen a range of independent initiatives in text encoding. Widely used formats include the Aozora Bunko markup and various forms of Markdown and their derivatives. When annotation is not embedded directly in the text, RDF is also commonly employed. Although RDF statements can be expressed using <standOff>, doing so still requires a solid understanding of the underlying data model. Consequently, effective education in text encoding cannot be achieved by teaching TEI alone. It must also address the roles, affordances, and limitations of coexisting formats and make interoperability strategies explicit—particularly through data conversion between TEI and other representations, and through TEI-centered transformations that enable heterogeneous datasets to interoperate, even though generative AI can now help draft such conversions and workflows.

Against this background, and with attention to the distinctive features of EAJ texts as they bear on encoding design, we are developing a TEI textbook that integrates TEI instruction with conversion-oriented pedagogy across multiple encoding ecosystems. This presentation reports the textbook’s scope and structure and situates it within these institutional and technological shifts.

A classic question to ask of any piece of hardware with a screen (and quite a few without) is, “but can it run DOOM?” – a question to which the answer is almost always yes. Though this paper isn’t quite so ambitious as to propose a method of running DOOM itself within a TEI-encoded document (yet), it does explore the ways in which TEI encoding can help us understand the broader field of DOOM’s vast network of mods, modders, toolmakers, and creators.

This initial exploration discusses possibilities for encoding metadata and some limited level information for the more than 14,000 mods for DOOM and DOOM II catalogued by Doomworld.com and the /idgames archive. Beyond analyzing how to adapt TEI standards to this particular kind of born-digital materials, this project is interested in tracing the evolution of DOOM modding over more than thirty years. What tools have modders used and created, and how has their usage shifted with new technological developments? Who makes mods, and how do the connections between individual modders and modding teams evolve? What do modders’ readme files tell us about their motivations and goals? How permeable is the boundary between a “mod” and an entirely novel game? This project seeks to investigate all these questions and more, and this presentation will serve as a first look at our findings.

The Voynich Manuscript remains one of the most enigmatic and undeciphered texts in the history of manuscript studies. Despite extensive scholarly attention, research has often been hindered by the manuscript’s undeciphered script, complex visual organization, and limited interoperability between textual and visual data. This project addresses these challenges through a digital humanities approach that re-encodes the manuscript’s herbal section within a structured TEI (Text Encoding Initiative) framework.

Using René Zandbergen’s widely used transliteration files as a foundation, this project transforms existing plaintext transliterations into a semantically structured TEI file designed for scholarly research. This project treats each illustrated plant as a discrete, encoded entity. By identifying and organizing plant illustrations within the TEI format, the project manages to systematically link transliterated text to corresponding botanical imagery while preserving the manuscript's unique glyphs and visual layout.

This structured encoding makes the Voynich Maniscript more machine-readable, searchable, and computationally accessible. Ultimately, this work demonstrates how careful TEI encoding can transform a famously opaque manuscript into a structured research environment, providing digital humanists, manuscript scholars, and students with new tools to explore one of history’s most persistent textual mysteries.