Margaret Jollie was born in Anstruther, Fife, Scotland on June 29, 1922. She entered nursing training at Cameron Bridge Hospital in Fife in September 1939, and on completing her training she nursed at Bridge of Earn Emergency Medical Services Hospital throughout WWII. There she nursed injured Allied servicemen, but also some Italian and later German POWs. During that time she kept autograph books in which she invited patients and other nurses to make entries. For years, her son Martyn planned to make a digital archive of the autograph book entries where descendants of those who submitted autographs might find information on their ancestors.

In December of 2022, Martyn’s cancer diagnosis became the impetus to complete the project, and Martyn and three of his children, Kiera, Neil, and Kevin, began what is now known as the Nurse Jollie Autograph Book Project. In creating the Nurse Jollie Autograph Book Project, our goals were 1. to digitize Nurse Jollie’s autograph books, 2. to make the book easily accessible to others investigating family history and WWII, and 3. to work together on a digital humanities family history project that presented key challenges, as well as key insights into our lineage.

The methods included scanning the pages of the book using a personal printer/scanner, rotating and cropping images using a photo editor by PhotoWorks, and saving the files according to the Autograph book and page number (e.g., Lg Brown_1_Smith_edited.jpg) before uploading to a shared Google Drive. Next, the pages were transcribed into a Google Doc within the shared folder, and completed in order of page number. Once the transcriptions were complete, the scanned images and transcriptions were added to a WordPress site dedicated to the project.

In this presentation, Kiera and Martyn Obbard will present the Nurse Jollie Autograph Book Project and, reflecting on our initial goals for the project, share the challenges of translation–or untranslatability–that we encountered. First, we will articulate the challenge of transcribing handwriting from WWII, and how we dealt with names or words that we could not agree on, and how we crowd-sourced some translations from friends fluent in Italian and Polish. We will discuss the decisions we made in representing sensitive content relating to sketches of Indigenous peoples. Second, we will outline how we worked through the structure of the page while working with no resources and from a chemotherapy room–specifically, touching on the organization of the page to mirror the act of flipping through the book (or translate the book form from print to digital archive, following N. Katherine Hayles’ theory of media translation), the inclusion of metadata, and the inclusion of a Surnames List to make it easier for visitors to quickly identify pages written by their ancestors. We will also share some surprising historical insights, specifically related to how Polish casualties ended up at Bridge of Earn. Finally, we will discuss the reception of the project to-date, and our plans for the next iteration of the project.

Starting in 2022, the Collaboratory for Writing and Research on Culture (CWRC, University of Guelph) started developing, within the SSHRC-funded partnership Staging Better Futures/Mettre en scène de meilleurs avenirs (SBF/MSMA), a sustainable digital archive of existing and new approaches to teaching, learning, governance, and administration that would contribute to reimagining post-secondary theatre training in Canada. To do so, they created the SBF/MSMA Archiving Working Group, composed of CWRC Director Susan Brown, Assistant Director Mihaela Ilovan, Research Assistant Alice Hinchliffe, and SBF/MSMA General Co-Directors Jennifer Roberts-Smith and Nicole Nolette.

One of the main ways of maintaining the project’s commitments to inclusivity, diversity, and equity and embedding them within the digital infrastructure of SBF/MSMA’s bilingual archive has been by mobilizing Nolette’s (2015) concept of “playful translation.” Previously used to refer to Franco-Canadian theatre artists’ multilingual targeting of specific audiences based on their different linguistic abilities, playful translation recognizes (somewhat like Cassin’s “untranslatables”) that some resistance to translation into English might best promote equity for minoritized languages and language speakers (Paré 1992; Cronin 1998). Playful translation is harnessed within the SBF/MSMA Archiving Working Group to formulate policies regarding the development and maintenance of bilingual (French-English) infrastructure, taxonomy and lexicon.

For example, while SBF/MSMA does not refuse to rely on automated dictionaries and machine translation tools like DeepL Translate or Drupal user-made dictionaries, priority is given to manual original translation using the free open-source digital translation platform Weblate. Chosen for its accessibility features, Weblate requires neither a particular digital installation nor any coding experience to navigate, thus offering a certain ease of use for individuals outside the field of digital humanities.

To organize and filter an extensive bilingual bibliography of curated resources on theatre pedagogy, the SBF/MSMA Archival Working Group is developing two taxonomies of terms linked to the goals of the project and the concerns appearing in the growing bibliography.

These taxonomies also appear as a lexicon on the SBF/MSMA website, offering definitions of each term. The bilingualism of this lexicon (which presents individual languages separately) recognizes the incommensurability of certain concepts across languages and surfaces the distinctions among and alignment of particular conceptual terms or expressions. This means that the French definition of a term or expression might not coincide with the English definition, that one might be longer than the other, that each might include different references.

In this paper, we propose a discussion on playful translation, linguistic minoritization and the development of a taxonomy and lexicon for SBF/MSMA. In resisting or refusing “translation” (in a one for one process where one is the original and one is the translation), are we excluding anyone unfamiliar with either French or English? What are the new spaces or discovery, limitations, and challenges to and of such “playful translation” processes?

Problématique

Comment transformer un corpus ancien, historiquement réservé à un cercle de spécialistes, en objet culturel partageable sans en aplatir la complexité? Cette communication propose une rétrospective sur onze années du projet Anthologia Graeca, une édition numérique ouverte et collaborative de la collection d’épigrammes grecques connue sous le nom d’Anthologie grecque. Loin d'être un « produit fini », l'objectif est de faire de cette longévité un instrument d’analyse : une durée suffisante pour observer ce que nos modèles, nos interfaces et nos promesses d’ouverture gagnent, perdent ou déforment lorsqu’ils rencontrent la matérialité des sources et la pratique quotidienne de l’édition.

Le projet constitue un terrain d’épreuve particulièrement exigeant par ses états textuels multiples, ses lacunes, ses scholies, ses séries thématiques et son ancrage constant dans des témoins manuscrits. Plutôt que de normaliser ces couches, l’infrastructure les traite comme des objets explicites, interreliés et traçables. L’enjeu n’est donc pas seulement de convertir un texte en données, mais de rendre visibles les opérations qui structurent sa circulation numérique (sélection, indexation, annotation, mise en relation, publication) et d’assumer l’écart inévitable entre l’objet philologique et ses représentations calculables.

Méthodologie

Le corpus rassemble plus de 4000 épigrammes, liées à des milliers d’entités (personnes, lieux, thèmes, topoï), ainsi que des scholies, des traductions et des liens vers des images du manuscrit Palatinus 23. La stratégie adoptée privilégie une « petite donnée dense ». Au lieu de viser l’accumulation, elle investit dans des notices riches, interopérables et réassemblables en parcours de lecture, en usages pédagogiques et en expérimentations computationnelles.

Sur le plan technique, la plateforme est une application web dotée d’une API REST documentée. Les données circulent sous forme d’enregistrements JSON conformes à un schéma public, ancrés par des identifiants canoniques et des URN CTS pour la citation à long terme, et enrichis par des liens d’autorité (e.g. vers Wikidata) pour favoriser la réutilisation et l’intégration dans un écosystème de données liées. Le flux de travail éditorial, versionné, permet à des étudiant·e·s, auxiliaires de recherche et chercheur·e·s confirmé·e·s de créer et réviser des notices, d’aligner des autorités, de lier scholies et images et de valider des résultats.

Conclusion

La contribution principale de cette rétrospective est de montrer ce que le temps long rend enfin visible: les ajustements du schéma au contact des cas limites, les compromis d’interface entre lisibilité et complexité, la manière dont une communauté s’approprie (ou contourne) les catégories proposées, et les coûts réels de la maintenance. La communication met également au premier plan les questions de pérennisation, tels la stabilité des identifiants, les migrations de formats, la documentation des transformations et la conservation des versions. L'Anthologia Graeca est présentée comme un laboratoire où l’édition n’est pas seulement un résultat, mais une pratique soutenue par des infrastructures durables, capables de garder lisibles et pérenne les choix, les incertitudes et les médiations qui font la rigueur philologique.