a three-day, in-person training event on how to use Wikisource, Transkribus, and related open digital humanities, infrastructures, methods, and tools to create, improve, publish, and reuse texts from cultural heritage collections.
The programme responds to a growing need for practical and sustainable ways of working with historical texts and dictionaries. It presents Wikisource not only as a transcription free and open-access platform, but also as a useful environment for research, teaching, and public engagement. Beyond powerful OCR models, Wikisource fully integrates the Handwritten Text Recognition (HTR) tool Transkribus, which has an agreement with Wikimedia to provide free access within the Wikimedia ecosystem.
A special feature of the school is its focus on dictionaries and other documentary heritage, especially in Portuguese, Latin, and other European languages. Participants will take part in hands-on activities such as transcription, proofreading, text structuring, editorial practice, metadata work, and examples of reuse for corpora and dictionaries.
The training brings together digital humanities, lexicography, scholarly editing, metadata, language resources, and cultural heritage. It is aimed at researchers, early-career scholars, librarians, archivists, museum professionals, educators, language specialists, and Wikimedia practitioners.
By the end of the school, participants will know how to transcribe and proofread texts in Wikisource, use basic editorial and metadata methods, understand how Wikisource can support research and teaching, and plan small reuse projects for cultural heritage and dictionary materials. Participants will also learn how to integrate citizen-science elements around Wikisource projects, turning transcription and proofreading into welcoming public-engagement activities that create real scholarly value.
- MA and PhD students in the arts, humanities, and social sciences;
Early-stage researchers and postdoctoral researchers in Digital Humanities;
- Librarians, archivists, museum and other GLAM professionals;
Lexicographers, terminologists, editors, and language resource specialists;
- Higher-education teachers and trainers interested in open textual heritage and collaborative digital methods;
- DARIAH and Wikimedia community members seeking stronger links with research and cultural heritage contexts.
No prior experience with Wikisource and Transkribus is required. The school has been designed to accommodate participants with different levels of digital expertise, combining introductory pathways with more advanced practical activities.
is a French Wikimedian with hands-on GLAM experience as Wikimedian-in-residence at the Musée de Bretagne (Les Champs Libres, Rennes), the heritage library and the university library of Clermont Auvergne. In that role, he worked at the interface between the institutions' teams and the Wikimedia community, helping to set up sustainable projects to publish and improve cultural heritage content across Wikimedia platforms, including Wikisource, with a focus on workflows, community collaboration, and quality assurance through proofreading.
is an Assistant Professor at the Faculty of Arts and Humanities of the University of Porto (FLUP) and a researcher at CLUP. She works in lexicography, terminology, and digital lexical resources, including TEI-based modelling (e.g., TEI Lex-0) and collaborative approaches to dictionary data. She has been involved in DARIAH WGs such as DHwiki, and has contributed to standardisation work in the lexicographic domain.
works in the field of open knowledge and free culture, with long-running involvement in Wikimedia projects (including training, outreach, and technical work). He is a founder of Wikimedia Portugal and has served in leadership roles there. He has been active as a Wikimedia editor since 2008, with interests that include the technical side of Wikimedia projects (e.g., automation/bots and data handling). He is associated with FLUP and a member of DARIAH WG DHwiki.
is a senior researcher at CHAM (NOVA FCSH / Universidade NOVA de Lisboa) who works on making large handwritten archives accessible through AI-based handwritten text recognition (HTR). He coordinated the TraPrInq project on Portuguese Inquisition trial records, using Transkribus to train and apply recognition models (including the “Portuguese Handwriting 16th–19th c.” model) and to integrate machine output with editorial protocols and human review. In this activity, he brings practical experience on how to use HTR to accelerate transcription, while managing its limitations, risks, and quality-assurance needs for reliable cultural-heritage reuse.
Luís Trigo (orcid: 0000-0002-3772-7081)
- CODA, CLUP, FLUP, DHwiki
Vera Moitinho de Almeida (orcid: 0000-0003-4979-8247)
- CODA, CITCEM, FLUP, INESCC, DHwiki
Alice Santiago Faria (orcid: 0000-0002-5006-4067)
- CHAM, NOVA FCSH, DHwiki
André Barbosa (orcid: 0009-0001-6866-6189)
- Wikimedia Portugal, DHwiki
The first day introduces Wikisource from the perspective of galleries, libraries, archives, and museums (GLAM), focusing on practical and sustainable workflows for publishing cultural heritage texts. Drawing on experience at the Musée de Bretagne and Clermont-Ferrand heritage library, participants will explore project setup, community collaboration, quality assurance through proofreading, and ways to connect Wikisource work with broader open-access and public-engagement strategies.
The second day focuses on dictionaries as a key use case for Wikisource. Using selected lexicographic and documentary materials, participants will learn how to move from digitised dictionary pages to proofread, structured and reusable text. The day will combine short methodological introductions with hands-on work on transcription, proofreading, entry structure, editorial conventions, source description, categories and basic links between Wikisource and related Wikimedia environments. Nicolas Vigneron will contribute examples from Wikisource and GLAM workflows, helping participants understand how collaborative proofreading, quality control and metadata practices can support sustainable lexicographic projects. The day will end with a practical reuse exercise for corpora, digital dictionaries, research or teaching.
The third day introduces AI-based approaches to handwritten text recognition (HTR) as a complement to Wikisource transcription and editorial workflows. Participants will discuss how recognition models can accelerate work with manuscript sources using Transkribus, which is embedded in Wikisource, their limitations and risks, and how to integrate them responsibly with human review, documentation, and scholarly editing practices to ensure reliable outputs for cultural heritage reuse.
